Intense pulsed light treatment in meibomian gland dysfunction_ A concise review

PURPOSE
To review the published literature related to application of intense pulsed light (IPL) for treating meibomian gland dysfunction (MGD).


METHODS
The literature search included the PubMed database and used the keywords "Intense Pulsed Light and Meibomian Gland Dysfunction".


RESULTS
IPL is a new instrumental treatment modality for MGD. This treatment modality was originally developed for use in dermatology and was later adopted in ophthalmology for treating MGD. IPL therapy for MGD can improve tear film stability, meibomian gland functionality, as well as subjective feeling of ocular dryness. However, in the reviewed literature, there was great variability in patient selection, evaluation criteria, and treatment protocols and durations.


CONCLUSION
Numerous studies report that IPL is effective for treating MGD and a safe procedure. There is great potential for further improvements to the procedure, as large comparative studies employing different treatment modalities are lacking.


Introduction
Dry eye disease (DED), also known as dry eye syndrome, dysfunctional tear syndrome, and keratoconjunctivitis sicca, is one of the most common ocular conditions prompting patients to seek eye care. It is prevalent in 5-50% of the population, depending on the diagnostic criteria and regions of the world; this multifactorial disease can occur at any age, but the elderly and women are more susceptible. The symptoms can vary, but in general manifest as ocular fatigue, dryness, discomfort, foreign body sensation, stinging, and blurred vision that interfere with daily function and quality of life. Therefore, DED imposes a substantial economic burden on patients and society owing to the considerable loss of work productivity and the use of medical resources [1][2][3].
Meibomian gland dysfunction (MGD) is considered to be the major cause of DED [4][5][6]. This chronic and diffuse abnormality is generally characterized by terminal duct obstruction and/or qualitative/quantitative changes in the glandular secretions [7]. The traditional treatment options for MGD, such as warm compresses, enhanced eyelid hygiene by removing blocked meibum, anti-inflammatory agents, antibiotics, and supplements with omega-3 fatty acids, do not achieve long-term satisfactory results. Therefore, the exploration of new potential therapeutic interventions has become a necessity [8][9][10][11].
Intense pulsed light (IPL), originally developed for use in dermatology, was introduced for treating MGD in 2015 [12]. Since then, attempts have been made to refine this new approach. The IPL device, also referred to as flashlamp therapy, is a light-emitting system that irradiates filtered polychromatic broad-bandwidth wavelengths with varying pulse durations for selective thermal damage of the target [13]. Several studies have reported the safe application of this procedure without any adverse effects, provided proper eye protection is employed [14][15][16][17][18][19]. Although the exact mechanisms underlying its https://doi.org/10.1016/j.jtos.2020.06.002 Received 20 January 2020; Received in revised form 18 May 2020; Accepted 4 June 2020 beneficial effects are complex and still not well understood, it has been suggested that it reduces telangiectasia, eradicates Demodex mites, results in thermal softening and liquefaction of meibum, modulates the secretion of pro-and anti-inflammatory molecules, and suppresses matrix metalloproteinases [11,[20][21][22][23][24].
This paper aims to review the status of IPL application in the treatment of MGD. The main question addressed is whether IPL as a stand-alone instrumental treatment or-concomitant with manual expression of the meibomian glands-can improve any of the ocular surface parameters.

Dry eye disease
DED can be caused by qualitative or quantitative alteration in the tear film. The tear film is structurally composed of three layers: (a) An outer lipid layer at the air surface produced by the meibomian glands that reduces evaporation of the aqueous layer; (b) an intermediate aqueous layer produced by lacrimal glands that provides nutrients, suitable osmolarity, and antimicrobial proteins; (c) an inner mucus layer on the epithelial surface produced by goblet cells for ocular surface coating and lowering epithelial cell hydrophobicity [25][26][27][28][29][30][31][32][33][34][35] (Fig. 1).
The Tear Film & Ocular Surface Society Dry Eye Workshop II report categorized DED into two major groups: aqueous-deficient dry eye (ADDE) and evaporative dry eye (EDE). The former refers to failure of lacrimal secretion and is mostly caused by Sjögren and non-Sjögren lacrimal disease. The latter refers to excessive tear loss from the ocular surface and results mostly from MGD. However, many cases present with mixed forms of DED [3,36,37]. The contribution of ADDE to DED incidence is approximately 10%, whereas more than 80% of cases are caused by MGD as well as the combination of ADDE and EDE [4].

Meibomian gland dysfunction
The meibomian glands, located in the tarsal plates of the upper and lower eyelids, are responsible for producing lipids and proteins (Fig. 2). Following muscular contraction of the lids, meibum is released onto the tear film, as these sebaceous glands terminal excretory ducts open at the posterior lid margin. Normal meibum reduces evaporation, affects tear film stability, and protects the ocular surface against microorganisms [26,57,58].
The two main subcategories for pathophysiology-based classification of MGD are low-delivery (hyposecretory MGD) and high-delivery (hypersecretory MGD). The occurrence of the former is linked to the hyposecretion or obstruction of the glands, which is believed to be the main reason for the development of MGD [7]. The latter has mostly been linked to seborrheic dermatitis. MGD can be caused mainly by obstruction of the terminal duct due to thickened opaque meibum or hyperkeratinization of the ductal system.

Biophysical principles
The emission wavelength of IPL discharged from flashlamps, usually ranging 500-1200 nm, is within the visible light and infrared radiation wavelength of the electromagnetic spectrum (Fig. 3). The produced broad wavelength is considered advantageous, as it can be absorbed by a variety of chromophores in human skin, for example, melanin (400-750 nm) and hemoglobin (578 nm), to develop heat. Both the absorption behavior and penetration depth of the light are functions of wavelength, and the latter is related positively to it (Fig. 4). Depending on the type and condition of the patient's skin, the application of a specific cut-off filter would help to selectively emit the optimal wavelength spectrum to target the structure at a particular depth [59][60][61][62]. For example, the intensity of light therapy is determined by the Fitzpatrick scale for scoring patients' skin types to minimize the risk of melanin damage and subsequent hypopigmentation [11,20]. In addition to the selection of the correct wavelengths, the efficacy of the treatment relies on the duration, intervals, and fluence of pulses applied to the target surface. The pulse delivery can be repeated singly, doubly, or triply for a duration commonly between 0.5 and > 20 ms [23,60,63].

Historical background
In 1976, Mühlbauer et al. [64] were the first to apply polychromatic infrared light as thermocoagulation for treating capillary hemangiomas and port-wine stains, leaving intact the non-affected overlying skin layers. In 1983, Anderson and Parrish [65] described the principle of selective photothermolysis through the application of pulsed radiation to damage pigmented structures, cells, and organelles. They demonstrated target selectivity via hemodynamic, histological, and ultrastructural responses in damaged as well as intact structures after in vivo irradiation. Despite such development, far too little attention was paid to the use of polychromatic light therapy before the 1990s [59,[66][67][68]. In 1990, Goldman and Eckhouse introduced a new high-intensity flashlamp for treating vascular irregularities of the skin. As a result, the first commercially available medical device with IPL technology, Pho-toDerm VL (Lumenis Ltd., Yokneam, Israel), was released to the market in 1994 [62,69]. In 2002, Rolando Toyos presented a case report of a patient treated with IPL for facial rosacea. The beneficial effects from the energy pulse were not limited to decreased facial erythema with dry eye, but were also accompanied by improvement in the clinical signs and symptoms of MGD. His collaboration with DermaMed Solutions led to the development of an IPL system that has opened new avenues for managing patients with DED [23,70]. As of yet, DermaMed has FDA labelling for treatment of rosacea patients, but not DED or MGD.

Device development
Since the 1990s, attempts have been made to modify Goldman and Eckhouse's original design for ease of operation, promoting safety and broadening the spectrum of potential indications. Early versions of the IPL device discharged the infrared portion of the wavelength spectrum, causing widespread epithelial damage, whereas in the later-generation IPL devices, the risk of adverse effects was minimized effectively using water as a filter. The modern generations of this device are equipped with computer-controlled capacitor banks and high-output flashlamps to emit polychromatic, noncoherent, and noncollimated light. In practice, the conversion of electrical to optical energy occurs when stored electrical current in the capacitor bank passes through a xenon gas-   filled chamber to produce bright light. Then, it is transferred through the distal end of the handpiece to be delivered to the skin surface using a sapphire or quartz block (Fig. 5). An ideal IPL device should have a large capacitor bank for supplying a constant current and emit wavelengths above 950 nm using water as an energy absorbent for excluding epidermal heating consequences [63,69]. Most of the available devices utilize low-intensity energies (8.5-20 J/cm 2 ) and may be accompanied with computerized software programs to provide users with default parameters for the treatment session. When managing patients with MGD, the proprietary "dry eye mode" setup must be used [11,71]. Before applying IPL (Fig. 6), the area being treated must be cleaned of makeup/facial care products. The eyes should be covered with protective eye shields to avoid intraocular damage. Moreover, to prevent the risk of epidermal burns due to close contact with the conduction crystal of the handpiece, an appropriate approach must be implemented depending on the system in use, such as integrated cooling of the IPL crystal (contact cooling, cryogen spray, or forced refrigerated air) and/or applying a thick layer of cold/ultrasound gel [11,20,60,63].

Pros and cons
The basic principle of IPL treatment is selective photothermolysis. It is often concomitant with adverse effects such as stinging pain, swelling, erythema, blistering, crusting, and scattered lesions. The possibility of nonspecific thermal damage to the surrounding tissue is prevented by reducing the pulse duration to lower than that of the thermal relaxation times. In addition, the improved models of IPL devices with the newest progressive set of parameters have significantly reduced the occurrence of acute adverse effects [60]. Handling and operation require a high level of training and experience, which casts doubt on the idea of home-use IPL devices as over-exposure and operator errors may cause ocular damage.
The price of a flashlamp device is lower than that of other light therapies such as laser, but the application is more often aimed at gaining and maintaining satisfactory results. Besides, IPL can cover a large treatment area and the technology is robust. However, the size and weight of the handpiece, which is in close contact with the skin, are inconvenient, and its flat surface hampers the treatment of irregular surfaces. The inability to focus the light as well as inconsistency in output (spectrum and fluence) are also important concerns [13,69].

Selection method
IPL is a widely used technology in dermatology. As IPL is a relatively new treatment modality for MGD, there are limited reports in the literature. For this review, we searched the PubMed database with the following search terms: "Intense pulsed light, meibomian gland dysfunction". The search was performed in December 2019 and returned 41 results. We noted a significant increase in the number of publications on IPL over the last 4 years (Fig. 7). The exclusion criteria for this review were: a) abstracts (1), b) letters to the editor (3), c) articles in other languages than English (3), d) review articles (8) and e) full text not available (2). Following these exclusions, 25 articles remained, whereof 17 were prospective and six were retrospective studies (Table 1).

Efficacy
Based on the studies included in this review, the IPL therapy procedures were performed either as a stand-alone instrumental treatment or in combination with manual expression of the meibomian glands (MGX). Thus, this review is divided into two parts. First, it summarizes the findings of 14 out of 25 reports on IPL therapy alone, followed by that of combination therapy (IPL with MGX). Prospective randomized studies are first reported, followed by prospective non-randomized and retrospective studies.

IPL as a stand-alone instrumental treatment modality
Fourteen [14,17,19,[72][73][74][75][76][77]81,[86][87][88][89] out of 25 stand-alone instrumental treatment modality studies, five were randomized [17,19,[72][73][74], four prospective and four retrospective studies. One article [84] did not mention the study type.     [72]. The treatment included three sessions on day 1, 15, and 45. IPL-treated eyes showed statistically significant improvements in lipid layer grade, noninvasive tear film break-up time (NIBUT), and subjective self-reported rating of dry eye symptoms. The tear evaporation rate and tear osmolarity did not differ between pre-and post-treatment sessions. The authors applied IPL treatment with a white-blocking filter in the placebo eye. Therefore, the placebo effect may have been compromised by light escaping the filter. The amount of light escaping the filter was not possible to quantify, thus, this may have contributed to control eyes receiving IPL to some degree. Another limitation in this study was the short follow-up time (45 days). Lasting effect of the treatment was not measured.
In another study with a larger sample size by Piyacomn et al., 114 patients with MGD were recruited and underwent the IPL treatment. They reported early improvement of OSDI score and meibomian gland functionality in the treatment group (day 15 vs. day 45) [19]. The authors reported a significant decrease in interleukin (IL)-1 receptor agonist in the group receiving IPL as well as the control group, but no differences between them. This might be a result of the inclusion of conservative measures such as warm compresses, lid scrubs and artificial tears in both groups. The authors found no differences in IL-6 level in either of the groups. At the 6-month follow-up, the treatment group had better OSDI scores, TBUT and meibomian gland functionality.
A study by Liu et al. [73] investigated change in inflammatory markers of the tear film of 44 patients with MGD following three IPL therapy sessions (4 weeks apart). The results showed a decline in IL-17A, IL-6, and prostaglandin (PG)E2 levels. In another study by Gao et al., the anti-inflammatory effect of IPL was compared with tobramycin/dexamethasone plus warm compress in 82 patients with MGD [74]. The authors concluded that IPL therapy led to improved TBUT and meibomian gland expressibility. The inflammatory cytokines IL-17A and IL-1β showed a transient downregulation at the 1-week follow-up of the IPL group, which became insignificant at 1-month follow-up. It is noteworthy that the clinical effects of IPL peaked at the 1-month follow-up. None of the studies reported adverse events related to the IPL therapy.

Prospective, non-randomized studies
In 2016, Jiang and co-workers carried out a prospective open-label study on the safety and effectiveness of IPL including 40 patients with MGD [14]. The study period was 75 days, with patients receiving four separate IPL sessions on day 1, 15, 45, and 75. The study demonstrated improvements in the subjective feeling of ocular dryness, tear film break-up time (TBUT), corneal staining, conjunctival injection, tear meniscus height, and meibomian gland functionality over the 75-day period.
Two years later, Yin et al. analyzed changes in meibomian gland morphology along with other clinical tests in a controlled cohort study of 18 patients with MGD [76]. The patients underwent three sessions with 30-day intervals, whereas the control group (n = 17) received eyelid hygiene. While OSDI score, TBUT, and meibomian gland functionality improved in both groups, only the IPL-treated group showed significant improvements in meibomian gland morphology.
Ahmed et al. assessed the effect of a single session of IPL therapy on both tear proteins and lipids in 24 patients with MGD [77]. The treatment resulted in elevated levels of tear lysozyme, lactoferrin, and albumin as well as total lipids, triglycerides, cholesterol, and phospholipids.
Similarly, a prospective study by Vigo and co-authors reported improved subjective symptoms, NIBUT, and lipid layer thickness in most patients following three sessions of IPL therapy over 45 days [78].

Retrospective studies
A multi-center cohort study by Gupta et al. involved 100 patients with MGD who underwent an average of four IPL treatments at 3-6 weeks intervals [86]. The authors reported decreased lid margin edema, facial telangiectasia, lid margin vascularity, and Ocular Surface Disease Index (OSDI) questionnaire score. There were also improvements in TBUT and meibomian gland functionality.
A retrospective case series by Karaca et al. involved 26 patients with mild to moderate MGD who underwent three IPL therapy sessions on day 1, 15, and 45 [87]. The patients showed improvements in Schirmer test, TBUT, OSDI and SPEED (Standard Patient Evaluation of Eye Dryness questionnaire) scores by day 45 without any adverse ocular effects. Another study published in 2019, with 19 patients using the same device and treatment protocol (three sessions on day 1, 15, and 45) reported improvements in NIBUT, lipid layer thickness, and subjective symptoms [82]. In contrast, OSDI score, tear osmolarity and meibomian gland loss remained unchanged.
Cheng and colleagues performed a retrospective study of 25 patients who underwent four IPL therapy sessions over the course of 42 days [89]. They reported a Demodex eradication rate of 20%, along with improvements in meibomian gland functionality and morphology, using meibography and in vivo confocal microscopy. These findings were corroborated in a single-blinded, randomized, controlled clinical trial performed by Zhang et al. [92]. This study compared the effect of IPL to a control group receiving tea tree oil (TTO), which is an established therapy for treatment of demodex mites. Both study groups showed a significant reduction of demodex infestation as well as improved clinical parameters. The cohort receiving IPL showed significantly improved OSDI, meibum quality and TBUT as compared to the TTO controls. Both treatments were similarly effective in Demodex eradication. However, TTO is reported to have eye irritation and redness as adverse effects [93]. Thus, the potential of IPL emerging as an alternative treatment of demodex should be a major avenue for future research. It is reported that Demodex causes inflammation [94]. Demodex infestations appear to affect over 80% of the population above the age of 60 [95]. As IPL can induce coagulation and necrosis of Demodex, it provides the rationale for using IPL to treat these infestations [92].

IPL therapy combined with manual expression of the meibomian glands
Eleven out of 25 articles reported results from the IPL therapy combined with manual expression of meibomian glands (MGX). Three studies had a prospective, randomized design, whereof six studies where non-randomized prospective and two were retrospective.

Randomized studies
Rong et al. carried out a prospective, randomized, double-masked controlled study with 44 patients [15]. The treatment group received three IPL therapy sessions at 4-week intervals. Besides, both eyes were treated with MGX and artificial tears. The group receiving IPL therapy demonstrated improved meibomian gland yielding secretion score (MGYSS), TBUT, SPEED, and corneal fluorescein staining (CFS). The control group, however, presented improvements only in the SPEED and corneal fluorescein staining scores. A few month later the same research group reported that IPL lacks long-term therapeutic effects (> 6 months) with regard to improved TBUT and meibomian gland functionality [75]. In their prospective, randomized, double-masked controlled study, 28 patients with MGD underwent three IPL therapy (real or sham) plus MGX sessions at 4-week intervals. The treatment group presented greater improvements in TBUT and MGYSS, which lasted until the 6-month follow-up, but there was no difference after 9 months. There was no difference in SPEED and CFS scores between the two groups.
In patients with refractory MGD [16]. The subjects received eight IPL therapy sessions with MGX. The IPL plus MGX group had significantly improved lipid layer thickness, NIBUT, TBUT, lid margin abnormalities, meibum grade, and SPEED score after 32 weeks. The control group did not receive sham IPL treatment and the study was not masked. It is therefore difficult to assess potential pre-and post-treatment evaluation bias.

Prospective, non-randomized studies
A prospective study by Dell et al. involving 40 patients (moderate to severe MGD) at two different sites showed improved TBUT, meibomian gland score, corneal staining, and SPEED score without changes in lipid layer thickness [79]. The treatment protocol was four IPL therapy sessions (3 weeks apart), followed by MGX of the upper and lower lids.
The following year, Albietz et al. used the same treatment protocol, except for a 2-week interval between session 1 and session 2 and a 4week interval between session 2 and 3 [80]. In that prospective study, the authors evaluated the effect of IPL with MGX in 26 patients with MGD. After 8 weeks' follow-up, they reported improved meibomian gland functionality, TBUT, corneal staining, lid margin, and limbal redness. The OSDI score was improved at the 12-week follow-up.
At the same time, Arita et al. carried out a prospective multicenter study assessing the effect of IPL plus MGX [81]. Thirty-one patients underwent 4-8 IPL plus MGX sessions with 3-week intervals. They reported improved SPEED, NIBUT, TBUT, meibum grade, lid margin abnormality scores, and conjunctival staining. Schirmer test values and meiboscore remained unchanged. In MGD, meibum in the meibomian glands becomes viscous. One of the mechanisms of action of IPL is to render meibum less viscous through liquification. This might be due to the thermal energy delivered to the skin surrounding the eyelids.
It is also believed that IPL therapy dilates the meibomian glands [12]. Therefore, the combination of IPL and manual expression of meibomian glands (MGX) is believed to yield better clinical results. Several studies have reported the outcomes of combination treatment involving IPL therapy and MGX.
Toyos and colleagues recently investigated changes in TBUT and the visual analog pain scale in patients with upper lid MGD after IPL therapy [18]. Their findings showed improved TBUT, global dryness scale, eye dryness (in the past 24 h), and frequency of ocular pain episodes. Using the same protocol, Seo and co-authors conducted a prospective case series study in 17 patients with rosacea-associated MGD [82]. Even though the patients had improved ocular surface [TBUT, noninvasive keratograph TBUT (NIKBUT), and staining score)] and meibomian gland parameters (lid margin vascularity, meibum expressibility and quality), the results obtained for TBUT, staining score and NIKBUT did not last more than 6-12 months after IPL treatment sessions.
Choi et al. studied the anti-inflammatory effect of IPL plus MGX on tear inflammatory markers along with subjective feeling of ocular dryness using the OSDI questionnaire, meibomian gland functionality, lid margin abnormality, TBUT, and ocular surface staining [83]. Thirty subjects who underwent three IPL plus MGX sessions at 3-week intervals were included in that prospective study. The treatment resulted in improved OSDI score, meibomian gland functionality, lid margin abnormality, TBUT, and ocular surface staining. Additionally, tear levels of IL-4, IL-6, IL-10, IL-17A, and tumor necrosis factor (TNF)-α were decreased after the treatment.

Retrospective studies
In a retrospective case series published in 2015 by Toyos et al. [12], patients with MGD (n = 91) received IPL therapy followed by MGX for 30 s with a cotton tip and the eye care practitioner's finger. The treatment protocol employed for improving DED included four sessions, with repeated IPL therapy every 30 days. Most of the patients (87%) had improved TBUT and self-reported dry eye symptoms (93%). Fourteen percent of the patients experienced adverse events such as blistering, cheek swelling, conjunctival cyst, floaters, hair loss at the brow and forehead, light sensitivity, and face redness. The authors concluded that the method was promising, warranting larger studies. A similar retrospective study by Vegunta et al., in 2015 included 35 retrospective cases who received 1-4 IPL therapy sessions with MGX every 4-6 weeks [85]. The authors concluded that 89% of the patients had reduced SPEED questionnaire scores and that 77% of the patients had improved meibomian gland function.

Safety outcomes
In 13 of the 25 studies considered in this review, no participant experienced adverse events. Ten studies did not report whether adverse events occurred during or after the IPL therapy sessions. Only two studies reported mild transient adverse events. Toyos et al. reported that two of 13 patients experienced adverse events [12]. The adverse effects included blistering, cheek swelling, conjunctival cyst, floaters, hair loss at the brow and forehead, light sensitivity, and face redness. Most of the adverse events resolved within 1 week. Rong et al. reported mild pain and burning during IPL therapy in five patients, while one patient had partial eyelash loss [15]. In contrast, six of the 25 studies concluded that IPL therapy is safe and without adverse events [14][15][16][17][18][19].
In the reviewed literature, the study outcomes share interesting similarities, such as consistent improvements in OSDI, TBUT, and meibomian gland functionality, and no changes in tear production (Schirmer test) and tear osmolarity (Tables 2 and 3). In contrast, post-IPL treatment changes in ocular surface staining are variable. Importantly, a large number of studies have shown that IPL is a safe treatment modality without any serious adverse effects. However, there is limited number of randomized clinical trials investigating the effects of the IPL therapy. Only eight of 25 articles in this review included randomized studies and only five of those studies were double-masked. Of note, the randomized studies were performed mainly in Asian countries (China, Japan and Thailand) and New Zealand. Research groups such as Toyos and Arita have pioneered in this field and published several articles, whereof three of eight randomized studies belong to those groups. IPL therapy is approved by the Food and Drug Administration (FDA) for treatment of rosacea and has been used as an off-label treatment for MGD, which may explain the lack of randomized studies in USA. However, in Europe, IPL therapy is used for this purpose by ophthalmic care personnel. Nevertheless, no randomized clinical trials have been reported by European specialists.
The comparisons of studies are limited by several factors. Data collection and inclusion/exclusion criteria were not identical between the studies. In addition, there was heterogeneity with regard to the study designs (retrospective/prospective). The questionnaires and clinical tests included in the IPL studies varied substantially. Several studies did not use common dry eye examinations. The most consistently reported analysis was TBUT. Moreover, there was great variability in the number of sessions, IPL therapy duration, as well as additional treatments, such as MGX, lid hygiene, warm compress, and lubricating drops (Table 1) This could be explained by the lack of evidence-based treatment guidelines for the use of IPL in MGD treatment. There is no consensus either with regard to the number of IPL treatment sessions or the session intervals. Further studies are warranted to reveal the optimal settings. The lack of standardized treatment protocol makes it challenging to directly compare results from IPL studies due to many unknown factors that potentially could contribute to either positive or negative clinical outcome. This obviously hampers the progress of future development of the technology. By standardizing the treatment protocol, it will be easier to determine if certain patients are more likely to respond positively to the treatment (sex, age, comorbidities, etc.). Moreover, it will ease the evaluation of the potential of combining IPL with other therapies to treat MGD. In-depth biochemical analyses of the tear film of patients with different subtypes of MGD using various omics technologies are likely to substantially advance our understanding of the mechanisms involved. This represents a second path to improved future therapy of MGD.
Although IPL therapy for MGD is considered both effective and safe, the application of this treatment modality is limited to certain patient groups. Subjects with Fitzpatrick skin type score > IV cannot receive IPL therapy as they are at higher risk of skin damage. Furthermore, IPL therapy is mainly applied to lower lids, avoiding penetration of the broad-spectrum light into intraocular tissues that can be potentially damaging. Nevertheless, positive effects in upper lid meibomian glands have been demonstrated even though the treatment was limited to the lower lids [86]. Lastly, the cost issue can be a limiting factor for some patients.

Conclusion
Based on the current literature, we conclude that IPL as a standalone treatment or in combination with MGX for treating MGD leads to the improvement of both symptoms and several ocular surface parameters. In addition, there is increasing evidence that IPL is a safe treatment modality for MGD. There is, however, a lack of high quality randomized clinical trials assessing IPL effectiveness. Larger, randomized controlled trials evaluating IPL effectiveness are required to explore the full potential of IPL. Moreover, future studies should widen + improved, --worsened, 0 -did not change, na -not applicable/not included, SRI -self-reported improvement, BR/LR -bulbar redness/limbal redness, CScorneal staining, CjS -conjunctival staining, CI -conjunctival injection, TMH -tear meniscus height, ST -Schirmer's test, MGM -meibomian gland morphology, TO -tear osmolarity, TER -tear evaporation rate, MGE -number of lower eyelid meibomian glands observed yielding liquid secretion, MGS -meibomian gland score, LMA -lid margin abnormality, LME -lid margin edema, LMV -lid margin vascularity, ME -meibum expressibility, MQ -meibum quality, MGYSS -Meibomian glands yielding secretion score. a Was not maintained at 6-and 12-months follow-up.  their inclusion criteria beyond MGD as IPL has proven to be effective for treating Demodex infestations, which is a highly prevalent condition in elderly patients. Finally, considering the safety profile of IPL, the potential should be investigated for its use on other, hitherto unexplored ocular surface conditions.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest
The authors except RA have no financial/non-financial competing interests. RA is a consultant of Lumenis, Japan.