Abstract
To evaluate the efficacy of low-level light therapy (LLLT) with near-infrared light-emitting diodes (LED-LLLT) for the treatment of dry eye. 40 patients were randomly assigned with a 1:1 allocation ratio to receive LED-LLLT (LLLT group, n = 20) or placebo treatment (placebo group, n = 20). Patients in the LLLT group received LLLT twice a week for 3 weeks, for a total of 6 treatment sessions. The primary endpoint was the changes in the fluorescein corneal staining (FCS) score. The secondary endpoints were the changes in the ocular surface disease index (OSDI) score, lissamine green conjunctival staining (LGCS) scores, tear film break-up time (TBUT), Schirmer test, and the meibomian gland dysfunction (MGD) index. These were evaluated before treatment and 4 weeks after start of treatment. The mean difference of score change in primary endpoint revealed significant improvement in the LLLT group, compared to the placebo. Among secondary endpoints, LGCS, Schirmer’s test, upper meibography scores showed significant improvements, while TBUT, lid debris, lid swelling, lid telangiectasia, meibomian gland secretion and expressibility scores had slight improvement without significant differences. No serious adverse events were observed. The use of LED-LLLT for the treatment of dry eye and MGD appears to be safe and beneficial.
Discussion
This study evaluated the effect of light-emitting diode-based LLLT (LED-LLLT) on a group of patients with dry eye disease and found positive results on improving symptoms. Dry eye patients often complain of blurriness and glare even though the result of visual acuity is normal18. The tear film breakup, tear hyperosmolarity, ocular surface inflammation, which causes an irregular tear surface, may be the cause of visual deterioration19. Meibomian gland dysfunction (MGD) has proved to be a risk factor for dry eye20. The gland dropouts with age, obstruction at the opening of duct, chronic damage, and poor quality meibum are associated with MGD21.
LLLT in general, or photobiomodulation, is a treatment approach currently used for dermatological and other medical purposes. The biological action of LLLT is thought to take place through the intracellular absorption of energy by the cellular membrane, intracellular organelles and molecules depending on the wavelength. Athermal and atraumatic cellular photoactivation occurs with light emitting diodes of specific wavelengths. This photoactivation is shown to repair damaged or compromised cells and improve cellular function in normal cells13. The chief medical indication for LLLT are in accelerating and enhancing tissue repair and promoting regeneration of a variety of tissues and nerves, reducing pain and inflammation, and preventing tissue damage22. LLLT in the present study comprised exposing the eyelid tissue to low-levels of yellow and near infrared light. The selection of laser parameters depends on the application target. The optical features of tissues differ markedly and are characterized by scattering and absorption coefficients, which depend on wavelength23.
Although LLLT has been historically performed using mostly laser-based light sources, the past decade has seen an increasing number of reports on LLLT delivered with light-emitting diode (LED) arrays (LED-LLLT). LED-LLLT offers advantages over laser diode (LD)-LLLT. Although LEDs are noncoherent, good quality LEDs are quasimonochromatic, with more than 98% of the photons at the rated wavelength, and the construction of the LED chips ensures good directionality with photons all travelling in approximately the same direction, albeit out of phase: this means that LEDs cannot be collimated or focused to a point. Secondly, because they are noncoherent and deliver a divergent beam of photon energy, LEDs are intrinsically safer than LDs because the pupil can gather only a minute fraction of the emitted noncoherent light, and as already mentioned LED energy cannot be focused to a point because LEDs are not a point source, which LDs are. Finally, LEDs can be mounted in planar arrays, thereby enabling irradiation of large areas of tissue in a hands-free manner, delivering a large area of homogeneous near-field irradiance and making precise positioning of the array over the target less necessary.
The skin chromophores (blood and melanin) possess high absorption bands at wavelengths shorter than 610 nm, therefore visible light does not penetrate deeply into the dermis. In the visible band from 610 to 700 nm, penetration increases by approximately 5 orders of magnitude, and at the wavelength of 830 nm, penetration is deepest before water becomes a major chromophore at wavelengths greater than 1150 nm. This results in the so-called optical window that covers the red and near infrared wavelengths and in which the effective tissue penetration of light is maximized24. Phototherapy utilizes light wavelengths between 390 and 1100 nm and may be pulsed or continuous wave. In standard condition, relatively low power densities (< 100 mW/cm2) are utilized25. To treat superficial tissue, wavelengths in the range of 390 nm to 600 nm are applied and longer wavelengths in the range of 600 nm to 1100 nm are applied to treat deeper tissues26. LLLT in the red to near infrared spectral range (630–1000 nm) and nonthermal power (less than 200 mW) is known to be effective for treatment of acne vulgaris27. A previous study evaluated the efficacy of a combination of 830 nm and 630 nm wavelengths in two sessions over 4 or 5 weeks to treat recalcitrant psoriasis employing LED irradiation and revealed no adverse side effects and a resolution of psoriasis28. A LED-based matrix module operating at the main wavelength of 830 nm, as utilized in our study, provided the effective penetration depth and was well tolerated on eye tissues14. The 590 nm component applied in the present study for the first minute of the total 11-min treatment time has as its targets the mother keratinocytes in the stratum basale in the epidermis, releasing adenosine triphosphate into the epidermal matrix and preconditioning the dermis. The subsequent 830 nm energy literally ‘washes out’ the clinical efficacy of the 590 nm energy at more than 3 orders of magnitude greater irradiance, and therefore targets all cells in the eyelids, including the meibomian glands. The combination of the effects of these wavelengths in the system used in the present study has proven highly effective in wound healing and skin rejuvenation, with particular efficacy accorded to the main wavelength of 830 nm29. With the combination of the inherent safety associated with noncoherent diffuse LED energy and the use of the metal contact eyeshields, no ocular safety issues were raised, and no serious adverse events occurred during treatment or the follow-up period.
In past few years, LLLT has been applied on dry eye patients to improve the symptoms and lower the severity of disease7,8,9. However, most of them were applied along with intense pulsed light (IPL) with positive effects, which makes it difficult to conclude the effect of LLLT itself. To our knowledge, this is the first randomized controlled trial to evaluate the sole effects of LED-LLLT on the treatment of dry eye disease30. Our study has shown a decrease in dry eye symptom, suggesting that LLLT can be used for lacrimal and meibomian gland alterations. LLLT is widely used by dermatologists, plastic surgeons and other specialties due to its analgesic, anti-inflammatory, and biostimulatory effects, and recent reports have shown the efficacy of a combination of LLLT with intense pulsed light (IPL) therapy for meibomian gland dysfunction, and LLLT on its own in the treatment of chalazia7,14,31. Evidence has therefore started to accrue for the safety and efficacy of LLLT on ophthalmic tissues, and we believe that our study has added to that evidence. Due to the fact that FCS reflects corneal damage and LGCS reflects conjunctival epithelium integrity, these improvements at week 4 in the present study revealed that LED-LLLT improved the ocular surface conditions. Schirmer’s test results in the LLLT group showed a significant improvement at week 4 compared to the placebo group. This could be the result of the well-proven anti-inflammatory properties of LLLT indirectly leading to tear secretion. Also, improvement in upper eyelid meibography was observed in the LLLT group. There were more meibomian gland dropouts in the lower eyelid than upper eyelid before the intervention, so it may be the reason why lower meibography score did not show inter-group difference. Since the area of upper eyelid is wider than lower eyelid, it seems that the photobiomodulatory effect of LLLT is efficient. LLLT delivered with a light-emitting diode array may be effective for dry eye syndrome by stimulating the function of both the lacrimal glands and meibomian glands6.
Our study as well revealed that there have been slight improvements in TBUT, lid debris, lid swelling, lid telangiectasia, meibomian gland quality, meibomian gland expressibility scores in the LLLT group compared to the placebo group, having greater mean difference of score changes after interventions. The tendency of these improvements can be correlated with the beneficial photobiomodulatory effects of particularly the 830 nm components. However, such slight difference between the groups could have been the effect of the concomitant use of sodium hyaluronate eye drop. Artificial tears are frequently used with anti-inflammatory or immunomodulatory ophthalmic solutions in clinical practice32.
The results of the current study revealed the positive effect of LED-LLLT on patients with dry eye, however there are still a few limitations in this study. First, the effect of LED-LLLT on dry eye with statistically significant differences were only observed among the mean differences of score changes in the primary and secondary endpoints. Upon random allocation of participants into two equal arms, some of the endpoint scores at baseline in the LLLT and placebo groups already showed small to medium effect sizes (0–0.676). Second, the current study confirmed the improvement of symptoms during the trial period, and follow-up results were not considered in analysis. Third, this study showed no significant differences between the improvement of symptoms and age or sex. Considering that the severity of MGD is related with the age and sex of patients, subsequent studies on a greater number of patients may be required.
In conclusion, this is the first study to investigate the efficacy and safety of LLLT with LEDs in patients with dry eye syndrome. Our study demonstrated that LED-LLLT at a dose of ≈ 60 J/cm2 per treatment session effected improvements in the signs and symptoms of dry eye. Therefore, our study revealed that LLLT applied as a treatment for dry eye can stimulate lacrimal gland and meibomian gland function. Such efficacy, in addition to the well-tolerated profile of LLLT, makes it a potentially useful treatment option for dry eye in clinical practice.
