Interferometry Evaluation of Precorneal Tear Film Lipid Layer After Intense Pulsed Light in Evaporative Dry Eye Disease Owing to Meibomian Gland Dysfunction: A Randomized, Single Masked, Sham-Controlled Study

Background: Inadequate meibomian glands (MGs) secretion can lead to dry eye signs and symptoms. Tear film lipid layer (TFLL) secreted by MGs protects and prevents rapid evaporation of tear film. Our purpose was to assess TFLL alteration and function in patients with evaporative dry eye (EDE) using tear interferometry after optimal pulse light technology (OPT) intense pulsed light (IPL). Methods: This prospective randomized examiner-masked sham- controlled study included 86 participants (142 eyes) with DED. IPL or sham procedure was performed on day 0, 21, and 42. Ocular Surface Disease Index (OSDI), non-invasive breakup time (NITBUT), interferometric fringe pattern determined TFLL quality, fluorescein staining (FS), and meibum gland (MG) were assessed at day 0, 21, 42 and 3-month. Results: At 3-month, TFLL, NITBUT, MG drop-out, MG quality, MG expressibility, FS and OSDI improved significantly (P<0.05) in the IPL group, while the sham group had no significant improvements. All DE parameters significantly correlated with the improvement in TFLL following IPL treatment. Additionally, artificial tears usage was significantly less in the IPL group from D-42 onwards. Conclusion: IPL treatment demonstrated the ability to improve TFLL quality and clinically reduced sign and symptoms of DED thereby reducing the frequency of artificial tears usage.

DE [27]. However, treatment discontinuation often leads to relapse of signs and symptoms DED as the positive effects of these treatments are not sustained for long period of time [28,29].
Intense pulsed light (IPL) treatment is commonly used for dermatological conditions and lesions [11]. IPL treatment utilizes a noncoherent polychromatic light source with wavelength spectrum of 500-1200 nm on the cutaneous facial sebaceous glands. This photothermal effect is thought to decrease inflammation and stimulate the meibomian glands. IPL treatment with or without MG expression has been proven to be an effective therapy for the improvement of signs and symptoms of DE due to MGD [11]. Although improvements after IPL treatment on the signs and symptoms of DE have been documented, the impact of IPL treatment upon precorneal tear film lipid layer has not been documented before. The purpose of this study is to assess the improvement in the TFLL and its subsequent impact upon TFLL interferometry patterns in patients with evaporative DE due to MGD.

Study design and participants
This study was conducted in compliance with the Institutional Review Board of He Eye Specialist Hospital, Shenyang, China in accordance with the tenets of the Declaration of Helsinki (approval number: IRB2019.K002.01). Consecutive subjects were recruited from He Eye Specialist Hospital, Shenyang outpatient department. Informed written consent was obtained from all participants after careful explanation of the nature and possible consequences of the study. Data from these participants was collected at the anterior segment department, a specialized ocular surface unit between January 2019 to January 2020.
Inclusion criteria comprised the following: (i) age ≥18 years, (ii) Fitzpatrick skin types 1 to 4, (iii) able and willing to comply with the treatment/follow-up schedule and requirements, (iv) diagnosis of DE based on (a) ocular symptoms, (b) non-invasive tear film breakup time (NITBUT) of ≤5 sec, and fluorescein staining score of ≥1 (on a scale of 0 to 9) according to the van Bijsterveld method [30]; (iv) visualisation of meibomian glands on each lower and upper meibography image, and (v) bilateral diagnosed at any stages of MGD, according to the analysis, which exceeded the sample size requirements for the preferred study power. Power calculations were conducted with NITBUT as the designated outcome, and showed that a minimum of 30 participants was required in the IPL treatment group and a minimum of 30 participants was required in the sham treatment group, to detect a clinically significant difference of 3-4 s in pair-wise comparisons, at 90% power (β = 0.2) and a two-sided statistical significance level of 5% (α = 0.05), with the SD of normal values being estimated to be approximately 5-7 s.
NITBUT: Keratograph 5M (Oculus, Germany) is a clinical instrument that uses an infrared light source of wavelength 880 nm to assess the ocular surface, tear film and meibomian glands (http://www. oculus.de/) [32]. During each assessment time point, non-invasive first tear film breakup time using the Keratograph 5M (Oculus, Germany) topographer was measured three times consecutively and the median value was recorded.
Meibography (Meibo-score): Keratograph 5M (Oculus, Germany) was used to capture the upper and lower eyelids were turned over and the meibomian glands were observed. Partial or complete loss of the meibomian glands was scored using the following grades (meibo-score) for each eyelid: grade 0, no loss of meibomian glands; grade 1, area loss was less than one third of the total meibomian gland area; grade 2, area loss was between one third and two thirds; grade 3, area loss was more than two thirds [33].  OSDI: Validated Chinese web-version of OSDI (Allergan Inc, Irvine, CA) was used to assess DE symptom frequency and the impact of these symptoms on vision-related function [34]. It contains 12 items, and the score can range from 0 (no symptoms) to 100 (severe symptoms) points; 0 to 12 represents normal, 13 to 22 represents mild DED, 23 to 32 represents moderate DED, and 33 or more represents severe DED.
Patient survey: Online question (How many times did you use preservative-free artificial tears yesterday?) was administered using a smartphone messaging platform to ascertain the frequency of artificial tears used during the past day by the participants in this study (e.g., once a day, twice a day and so on).

Treatment
The Toyos protocol was used in all treatments [35]. All patients had a minimum of 2 treatments, each separated by 3 weeks. IPL was performed using M22 IPL system with optimal pulse technology (OPT) (Lumenis Ltd., Yokneam, Israel). Its xenon lamp that emits IPL at 515-1200 nm and a 560-nm filter and OPT makes IPL pulses more stable and highly repeatable, therefore M22 IPL treatment is considered more effective in targeting meibomian glands [36].
The sapphire-cooled 6mm cylindrical light guide set at a fluence of 10 J/cm2 (Table 1) Participants were randomly assigned to one of two treatment groups and underwent IPL treatment with 12 homogeneously spaced pulsed light to both eyes or sham treatment to both eyes, at days 0, 21, and 42 by a non-masked trained clinician, who was not involved in data collection process of this study. Randomisation was managed by computer-generated random generously on the patient's targeted skin area. Each participant in the IPL treatment group received 12 pulses of light bilaterally (with slightly overlapping applications) from the preauricular area and across the cheek. While participants allocated to the sham treatment group followed the same pre-treatment protocol with opaque goggles and ultrasound gel application and then a non-active IPL device was placed on the periocular area and was moved 12 times to simulate treating different areas of the periocular area, while an active IPL device was fired 12 times in the same room to imitate the acoustics of an active IPL device.
All participants were requested not to wear facial makeup and contact lens on treatment day and not to wear CL 3 weeks before baseline measurements and for the entire course of the study. To prevent possible facial pigmentation secondary to IPL treatment, participants were informed to avoid direct sun exposure for 1 month after each IPL treatment.   (Figure 1).

Patient Characteristics
Baseline clinical DED parameters did not differ between the IPL treatment and sham group (all P > 0.05; Table 3).    Figure 2). Additionally, along with TFLL scores, NITBUT was also significantly prolonged in the IPL group compared to the sham group at all assessment time points (Table 3). Meibo-score which assesses the MG dropout, MG quality and expressiblity was significantly improved (P<0.05) in IPL group at day-21, day-42, and 3-month in comparison to sham group (Table 3) (Figure 3).

Correlation between TFLL score and other measurements
The correlation analyses between changes in standard deviation of tear film lipid layer   Data are considered statistically significant at * P< 0.05, ** P< 0.01, and *** P< 0.001.

Safety data
No systemic adverse event was observed during the study. Among all visits, BCVA did not change significantly. The maximum IOP recorded was 14 mmHg, and no significant change was recorded at day-21, day-42 and 3-month in comparison to baseline measurements and sham group. ECC in all visits among participants remained similar and showed no significant change. No depigmentation, blistering, swelling, redness, and hair loss at the brow, eyelash loss, or on the ocular surface were observed after the IPL treatments (Table 5). IPL treatment has been extensively used in dermatology in treating acute and chronic dermal inflammatory disorders [33]. As documented in previous studies, photobiomodulation effect of IPL treatment can be safely used to improve the meibomian gland function and thereby stabilize the tear film and reduce ocular surface inflammation in evapourative DED [33,37].
Photobiomodulation is light-induced photochemical reactions at various biological scales by laser, LED, broadband, visible and near-infrared light, including IPL [37,38]. This process involves photons penetrating tissue and interacts with chromophores located in cells that leads to photophysical and photochemical changes and alters changes at molecular and cellular level [39]. IPL treatment has been documented to induce positive physiological reactions in diseased and damaged tissues to accelerate wound healing and tissue regeneration [40,41], increase circulation, reduce acute inflammation [42], and help restore normal cellular function [43].
The tear film lipid layer provides support for maintaining tear film homeostasis and also insight into the pathophysiology of DED [44]. Improving the quality of tear film lipid layer can therefore reduce evaporation of tears and enhance tear film stability as meibum has been reported to contain antimicrobial properties that keep the lid margin healthy [44]. While some researchers speculate that thermal energy transferred by IPL liquefies obstructed meibomian glands observed in MGD, relieving obstruction of glands and promoting the release of meibum.
However, IPL treatment for DED due to MGD is recommended not to be performed directly over the eyelids rather around the periorbital area surrounding the eyelids [28,45,46]. Therefore, IPL improves signs and symptoms of DED by selective ablation of superficial blood vessels by targeting chromophores in hemoglobulin and thereby reducing telangiectasias, erythema and reduction in inflammatory markers in the surrounding area of application [47] The results in our study are similar with other studies [48]. We found NITBUT after IPL treatment significantly increasing, as previously reported by other researches, showing an improvement of tear film stability. Because of increased meibum secretion and change in the viscosity and quality of meibum, the tear film becomes more stable, resulting in an improvement in dry eye symptoms. Garrrigue et al. showed lipid-based treatment is effective in improving the symptoms and signs of dry eye, making it a promising treatment option in the treatment of DED [29,46]. Similarly, in our study ΔTFLL showed a significant correlation with reduction in ΔAT. Ahmed et al. noted that significant improvements were observed in tear protein concentrations and molecular weight after IPL treatment that proved it improved tear protein and lipid content and composition [29,46]. Therefore, as the results shows ΔTFLL score significantly decreased (improved TFLL) significantly correlated with improvements in ΔNITBUT, ΔAT, and ΔOSDI score, elucidating that improved TFLL contributed to reduce signs and symptoms of DED. Sustained reduction in NITBUT and TFLL have been demonstrated to increased friction resulting in damaged corneal and bulbar conjunctival epithelium increasing [44,49] and various studies have reported improvements in ocular surface epithelial damage following IPL treatment [44,49]. While in our study FS and MG dropout did significantly improve at 3 months, ΔFS and ΔMG dropout didn't demonstrate significant correlation with ΔTFLL. It is possible that a longer follow-up is required for the improved ΔTFLL to significantly correlated with ΔFS and ΔMG since FS and MG dropout is a manifestation of chronic DED pathology.
There are several limitations in this study. Dry eye is a multifactorial disease and factors such as inflammatory markers and osmolarity were not included in this study. However, it can be speculated from the findings of other studies that these factors would possibly correlate with ΔTFLL [44,49]. Another limitation is the range of participants age for IPL and sham group was 21 to 34 years, therefore our findings cannot be generalized. Lastly, it was not possible to carry out an ideal sham group since they experienced the treatment without the OPT-IPL emitting light on their targeted area. Future studies will focus on larger sample size to optimise the power of the study and include a wider age range as DE is more prevalent in the elderly.

Conclusion
In summary, our findings suggest that OPT-IPL treatment significantly enhances TFLL and improvements in signs and symptoms of DED can be attributed to improved meibomian gland function.