FormalPara Key Summary Points

Why carry out this study?

The efficacy and safety of the intense pulsed light (IPL) device for specifically treating severe evaporative dry eye disease (DED) has not been determined.

The impact of IPL therapy on various ocular surface parameters in participants with severe evaporative DED has not been explored.

This study aimed to determine the safety and efficacy of IPL therapy in improving the signs and symptoms of evaporative DED and to perform a comparison with a sham control group.

What was learned from the study?

IPL can be safely used to manage signs and symptoms of evaporative DED.

A minimum of two IPL sessions were needed to observe significant improvements.

Introduction

Dry eye disease (DED) is multifactorial in etiology and characterized by ocular pain, visual disruption (blurred vision and hazy vision), tear film instability, and tear film hyperosmolarity, which may damage the ocular surface [1]. Some of the possible causes of DED are aging [2], menopause [3], meibomian gland dysfunction (MGD) [4], Sjögren syndrome [5], conjunctival fibrotic disease [6], refractive surgery [7], and systemic [8] or topical [9, 10] drugs. Insufficient tear production or excessive evaporation of the tear film has been documented to create a hyperconcentration or hyperosmolarity of the precorneal tear film, which disrupts tear film homeostasis [11]. Tear film hyperosmolarity can cause cell morphologic alterations, inflammatory cascades, cell death, tear film instability, and a further increase in hyperosmolarity [12]. MGD causes evaporative dry eye (EDE), the most frequent form of dry eye [11]. Anti-inflammatory medications, antibiotics, warm compresses, eyelid hygiene, and meibomian gland expression are MGD treatment standards [13]. MGD has been found to be linked to eyelid skin inflammatory disorders [14]. Rosacea affects 5.46% of the adult population (range 0.09–24.1%), with 58% having MGD [15]. Ocular symptoms precede cutaneous rosacea in 15–10% of instances, indicating a subclinical variant [14, 16].

IPL is a popular therapy for facial skin rosacea and over a decade ago, Toyos et al. [17] found that IPL theraoy for facial skin rosacea improved dry eye symptoms. In standard practice, IPL pulses are administered from tragus to tragus, below the lower eyelids and including the nose, in each session. Several studies have shown that IPL treatment decreases dry eye symptoms and indicators [18,19,20]. While previous studies have reported on the benefits of IPL therapy for a wide severity spectrum of DED, none have focused on severe DED, and there is a reasonable possibility that combining a wide severity spectrum of DED into single group (i.e., mild, moderate, severe) might give us an understanding of the entire group but not a specific sub-group (i.e., the severe dry eye group). Therefore, the aim of this study was to assess the treatment efficacy of IPL specifically for severe evaporative DED.

Methods

Study Design

This randomized control study was conducted by the Department of Ophthalmology, He Eye Specialist Hospital, Shenyang, China, between January and October 2021. The study protocol was approved by the Ethics Committee of He Eye Specialist Hospital (IRB2019. K002.01) and adhered to the tenets of the Declaration of Helsinki. Informed and signed consent was obtained from all participants before the treatment in this study.

Inclusion criteria comprised the following: (i) age ≥ 18 years, (ii) Fitzpatrick skin types I–IV, (iii) capable and willing to comply with the treatment and follow-up obligations, (iv) a determination of severe DED based on (a) the ocular surface disease index (OSDI; a score of ≥ 33 represents severe DED), (b) a non-invasive tear film breakup time (NITBUT) of ≤ 5 s, or a conjunctivocorneal staining score (CS) of ≥ 3 points according to the Asian Dry Eye Consensus [21].

Exclusion criteria: (i) existing ocular trauma, infectious diseases, recent surgical history; (ii) skin defects, pigmentation, moles, scars in the treatment area, skin cancer; (iii) autoimmune diseases, skin allergies; (iv) pregnancy or lactation; and (v) Fitzpatrick skin type V or VI.

Treatment

All treatments used the Toyos protocol and patients received three sessions separated by 3 weeks. The M22 IPL system (M22; Lumenis, Yokneam, Israel) is a new-generation IPL device with optimal pulse technology (OPT) and a modular laser multi-application platform that assures steady and predictable fluence throughout each flash. The wavelength of light emitted by the M22 OPT system varies from 515 to 1200 nm (nm) and is emitted by a xenon lamp. Adjusting a 560 nm filter on the M22 OPT system, appreciate energy for various skin tones can be achieved (range of 11–14 J/cm2). At day 0, day 21, and day 42, each patient completed a series of three IPL or sham therapy sessions (Fig. 1).

Fig. 1
figure 1

Study design

Full size image

Experimental Design

Participants were randomly allocated to one of two groups (IPL or sham) and received IPL treatment with 12 homogeneously spaced pulses of light to both eyes at days 0, day 21, and day 42, as administered by a non-masked qualified clinician who was not engaged in data collection. Computer-generated random numbers assigned successively enrolled individuals. The investigator who collected data at days 0 (baseline), 21 (week 3), 42 (week 6) and 82 (week 12) from all participants (IPL and control groups) was not aware of the participants’ therapy allocations. At day 0, day 21, day 42, and day 82, participants wore opaque goggles and ultrasonic gel was liberally administered to their targeted skin. The IPL therapy group received 12 bilateral light pulses to the periocular and cheek regions. In the sham therapy group, opaque goggles and ultrasound gel were applied, and a non-active IPL device was placed on the periocular region and moved 12 times to simulate treating different areas around the eyes. In the same room, an active IPL device was fired 12 times to simulate the acoustics of active IPL device therapy. Participants were instructed not to use any additional dry eye medications or eyedrops, including preservative-free artificial tears, throughout the course of this study (Fig. 1). All participants were requested to have their designated IPL or sham therapy session at day 0, day 21, and day 42; participants failing to do so were excluded from the study.

Treatment Procedure

Ultrasound gel was generously applied to the lower eyelids and preauricular region. Ultrasound gel was generously applied to the lower eyelids and preauricular region. Opaque goggles was worn by participlants while 12 light pulses were applied from the left preauricular area, over the cheekbones and nose, to the right preauricular area, reaching up to the inferior border of both eyes (regions of application overlapped somewhat). On each treatment day, the participants went makeup free and were advised to avoid direct sun exposure for 1 month following the IPL treatment to prevent face pigmentation.

Clinical Assessment

Before every OPT-IPL therapy, best corrected visual acuity (BCVA), OSDI, NITBUT, tear film lipid layer (TFLL), endothelial cell count (ECC), intraocular pressure (IOP), meibomian gland (MG), and CS assessments were performed in this order at day 0, day 21 (3 W), day 42 (6 W), and day 82 (12 W). An experienced doctor administered the IPL and sham therapy, and another trained doctor (blinded to participant allocation) gathered test data.

IOP was assessed using a non-contact tonometer (NT-510, NIDEK, Japan). ECC was assessed using a corneal endothelial counter (SP-3000P, TOPCON, Japan). NITBUT was assessed using the Keratograph 5M (Oculus, Germany) topographer. Three consecutive measurements were taken, and the median value was entered as the final reading.

TFLL quality was assessed noninvasively via TFLL interferometry using a DR-1 instrument (Kowa, Nagoya, Japan). The results were graded using the Yokoi dry eye (DE) severity grading system, where grade 1 corresponds to a somewhat gray color and uniform distribution, grade 2 corresponds to a somewhat gray color and non-uniform distribution, grade 3 corresponds to a few colors and a nonuniform distribution, and grade 4 corresponds to many colors and a nonuniform distribution

Meibography (MG Score). Keratograph 5M (Oculus, Germany) was used to capture the infrared images of the upper and lower eyelids after turning it over and exposing the meibomian glands. Partial or complete loss of the meibomian glands was scored for each eyelid using the following grades (MG Score): 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.

Meibomian gland function. The meibum quality and meibomian gland expressibility of the upper eyelid were assessed. (i) Meibum quality: eight meibomian glands in the middle parts of the eyelid were assessed using a scale of 0–3 for each gland: 0, clear; 1, cloudy; 2, cloudy and granular; and 3, thick (like toothpaste). (ii) MG expression: five meibomian glands in the middle part were evaluated on a scale of 0–3: 0, all glands expressible; 1, 3–4 glands expressible; 2, 1–2 glands expressible; and 3, no glands expressible. The average scores of these eight glands were calculated as the total score.

Using the methods of Arita et al. [22], the double vital staining test was used to assess conjunctival and corneal epithelium damage. The conjunctival sac was injected with 2 microliters of a preservative-free mixture containing 1% sodium fluorescein and 1% lissamine green. The eye was sectionalized into three equal sections representing the temporal conjunctiva, cornea, and nasal conjunctiva. A staining score (maximum: 3 points; minimum: 0 points) was assigned to each area. The CS was used to record the combined results from all three sections on a scale from 0 (normal) to 9 (severe) [23].

Statistical Analysis

The statistical analyses were conducted using SPSS statistics software (version 25.0; SPSS Inc., United States). The mean standard deviation (SD) was used to express descriptive statistics for continuous variables, whereas number (percentage) was used for binary variables. The chi-square test was used to analyze categorical data. Using the Kolmogorov–Smirnov test, the normality of the variables was determined. A linear mixed model with Bonferroni post-hoc analysis was used to evaluate repeated measurements of continuous variables, including IOP, ECC, BCVA, OSDI score, and NITBUT. Generalized linear mixed model analysis with Bonferroni post-hoc analysis was used for repeated measurements of discrete variables, including the TFLL, CS score, and MG assessments. A P value of ≤ 0.05 was considered to show statistical significance.

Results

Initially 30 participants were enrolled in each group. Two participants in the IPL group and 9 participants in the control group did not complete the study (Fig. 1). The final analysis included 28 (56 eyes) participants in the IPL group and 21 (42 eyes) participants in the control group. Table 1 displays the demographic information regarding treatments and the control group. Age, gender, duration of dry eye disease, and other ocular parameters were found to be similar when comparing the two groups (P > 0.05).

Table 1 Demographic data
Full size table

The mean OSDI score at baseline was recorded as 40.98 ± 7.29 and 42.02 ± 6.57 for the IPL and control groups, respectively (Fig. 2). Significant differences between the groups in the OSDI score were found at 3 W, 6 W, and 12 W (Table 2).

Fig. 2
figure 2

Comparison of the total mean OSDI between groups

Full size image
Table 2 Comparative assessment of the therapy and control groups
Full size table

The mean NITBUT at baseline assessment was 3.60 ± 0.59 and 3.79 ± 0.61 s for the IPL and control groups, respectively (P > 0.05) (Fig. 3). Significant differences between the groups were found at 3 W, 6 W, and 12 W (Table 2) (Fig. 3).

Fig. 3
figure 3

Comparison of mean NITBUT between groups

Full size image

The mean TFLL score was not statistically different between the IPL group and the control group at baseline (Fig. 4). A significant difference in mean TFLL score between the IPL group and the control group was found at W 6 and W 12 (Table 2).

Fig. 4
figure 4

Comparison of mean TFLL between groups

Full size image

Based on the Asian DE assessment system, the CS severity score increases from 0 to 9. At baseline, the mean CS score was 3.67 ± 1.34 and 3.71 ± 1.09 in the IPL group and the control group, respectively (Fig. 5). At 3 W, 6 W, and 12 W, the CS score improved to 3.61 ± 1.27, 3.46 ± 1.16, and 3.32 ± 1.11, respectively in the IPL group (Table 2).

Fig. 5
figure 5

Comparison of mean CS score between groups

Full size image

The mean MG Score at baseline was 1.93 ± 0.26 and 1.83 ± 0.38 in the IPL group and the control group, respectively (P > 0.05) (Fig. 6). Significant differences between the groups were found at 3 W, 6 W, and 12 W (Table 2).

Fig. 6
figure 6

Comparison of the mean MG Score between groups

Full size image

The mean MG quality score at baseline was 2.68 ± 0.07 and 2.55 ± 0.08 for the IPL and control groups (P > 0.05), respectively (Fig. 7). A significant difference between the groups was found at 6 W and 12 W (Table 2).

Fig. 7
figure 7

Comparison of the mean MG quality score between groups

Full size image

The mean MG expression score was found to be significantly improved in the IPL group at 3 W, 6 W, and 12 W when compared to the control group (Fig. 8) (Table 2).

Fig. 8
figure 8

Comparison of the mean MG expression score between groups

Full size image

During the course of the trial, no systemic adverse events were recorded. BCVA did not vary substantially between the baseline and all subsequent visits in either group. The mean IOP was not significantly different from baseline for both the IPL group and the control group. There was no significant difference between the IPL group and the control group with regards to ECC across all visits. No depigmentation, blistering, edema, ocular surface redness, hair loss, or eyelash loss was noted after IPL therapy (Table 3).

Table 3 Comparison of safety parameters between the therapy and control groups
Full size table

Discussion

While the mechanism of action of IPL for treating DED is not entirely understood, but the majority of authors suggest that this therapy results in the destruction of superficial blood vessels [24, 25], with a subsequent reduction in local inflammation [26,27,28], and antimicrobial [29], anti-inflammatory [28] effects. Currently, IPL therapy has been found to be beneficial for various forms of DED [18, 30,31,32]. Throughout their development, IPL devices have been changed to lessen the adverse effects of IPL and broaden its indications [33]. The newer versions permit quartz or sapphire crystals to filter the wavelength spectrum in the direction of longer wavelengths, which affect deeper skin structures relative to the epidermis [33]. The detachable filters also allow the selection of the wavelength required for certain lesions [34, 35]. In a similar fashion, the pulse energy has been fractionated, and novel cooling mechanisms have been developed to prevent skin harm [34, 35]. Among the possible adverse effects, discomfort during administration is the most common. This has been alleviated by the installation of modern refrigeration systems [35]. There have been reports of skin erythema and edema hours or (less often) days following therapy; however, both are transient. Pigmentation alterations and hypertrophic scars are seldom seen. Ocular problems have also been reported when using high fluences such as 20 J/cm2 [34]. There have been reports of anterior uveitis, iris transillumination abnormalities, and modifications that result in eye synechiae and pupillary obstruction followed by angle closure [36,37,38,39].

In our study there were no adverse events reported by the participants. At the 3-week assessment, among the seven dry eye parameters observed in our study (OSDI, NITBUT, TFLL, CS, MG Score, MG quality, and MG expression), CS and meibomian gland quality showed no significant difference between the IPL and control groups. However, at the 6-week and 12-week assessments, only one parameter (CS) did not show a significant improvement, suggesting that the benefits of IPL therapy for severe EDE can be realized after receiving at least two sessions. The major indicator of ocular surface injury is CS [40], which is considered a significant factor in the DED diagnostic process [41]. Numerous researchers have attempted to demonstrate the existence of a correlation between DED symptomatology severity and corneal staining in MGD participants. Some studies have found weak positive correlations between symptomatology severity and corneal staining [42], while others have demonstrated that there is no correlation between the two parameters [43,44,45]. Recently, Llorens-Quintana et al. [46] examined the relationship between MG irregularity, dry eye symptomatology, and several ocular surface indicators, such as corneal staining. They noted that there was no correlation between corneal staining and MG irregularity or MG loss. Xiao et al. [47] studied a sample of the two main MGD delivery categories—high delivery and low delivery—which can be segregated into four MGD states: hypersecretory MGD, undefined MGD, hyposecretory MGD, and obstructive MGD). They discovered that low-delivery groups (hyposecretory and obstructive MGD) had more severe dry eye symptoms and higher ocular surface staining scores than the high-delivery MGD groups. The current investigation revealed that while signs and symptoms of EDE can be treated with IPL therapy, a longer therapy course might be needed to improve the CS score.

This research has significant drawbacks. First, the age range of the individuals in our study did not include the elderly, so the results cannot be generalized to all the population, since DE is more frequent among the elderly. Second, this research did not include parameters such as inflammatory indicators and osmolarity, which contribute to dry eye. Nonetheless, based on the results of previous research, one might assume that these characteristics may be linked with TFLL [28, 42, 46, 48]. Lastly, a larger sample size is needed to improve the study's statistical power.

Conclusion

In summary, the findings of this current study suggest that OPT-IPL therapy significantly improves signs and symptoms of EDE following a minimum of two sessions of IPL therapy.