Intense Pulsed Light Combined With Meibomian Gland Massage for Recurrent Corneal Erosion: a Review of Clinical Effectiveness


 Purpose: To observe the therapeutic effect on recurrent corneal erosion syndrome (RCES) by intense pulsed light (IPL) combined with meibomian gland massage. Methods: We recruited 30 patients (30 eyes) with RCES as the RCES group and 31 patients (31 eyes) as the control group. Both the groups received eyelash sampling, Demodex count, meibomian gland infrared photography, and the blepharolipin score. The RCES group was divided into the treatment group and the observation group again. Results: Parameters in the RCES group were higher than those in the control group. The parameters of the treatment group were decreased compared with the observation group. During the follow-up period, one patient in the treatment group relapsed. Conclusion: Meibomian gland dysfunction and Demodex infection may be associated with RCES. IPL combined with meibomian gland massage can significantly improve meibomian gland function, reduce the number of Demodex, and effectively control the relapse of RCES.

5. Demodex examination of eyelashes [7] : pluck 3 thick eyelashes from the middle 1/3 of the upper and lower eyelid each eye (6 eyelashes in all), put them on the microslide, covered with a coverslip, add saline from one side, observe the Demodex under the photoelectric microscope, and count the number. Therapeutic Method All the patients were treated with antibiotic spongarion and arti cial tears eye drop after the rst-time consultancy. After two weeks of eye drop treatment, they were randomly divided into two groups, the patients in the treatment group treated with IPL combined with meibomian gland massage three times, once every 28 days; The patients in the observation group treated with preservative-free arti cial tears eye drop only.
We used SOLARI (Lutronic, Korea) intense pulse optical equipment for the treatment; parameters were tested in the patient's jaw before the rst treatment, adjust the IPL energy according to the patient's skin color, the energy range was 9 13 J/cm², kept the patient's skin clean and dry, and wore an eye mask to protect the patient's eyes. The treatment area was uniformly coated with a medical coupling agent for 5 ~ 10 mm. Doctor was expected to wear goggles during the whole treatment. Four therapeutic parts were selected from the left temple to the right, putting the probe on the coupling agent part, press the on-off button to emit the pulses of light, then move to the next part, and redo the operation until all the therapeutic parts have been done. When the treatment has been nished, remove the skin coupling gel, drop 0.4% bupivacaine hydrochloride (Santen Pharmaceutical co. LTD Japan) into the conjunctival sac two times, once every 5 minutes, then massaged the upper and lowered meibomian gland with meibomian gland massage tweezers. The same physician performed all procedures. When all treatment has been nished, ice the eyes for 5 minutes.

Therapeutic Evaluation
After the accomplishment of three treatments, Compared the meibomian gland loss score, blepharolipin score, the number of Demodex, and the disease recurrence during the one-year-follow-up period in the two groups respectively.

Statistical Analysis
SPSS22.0 (SPSS, USA) was used for statistical analysis. In this study, all counting data were skewed by the shapiro-wilk test and expressed by M (Q1, Q3).
The Mann-Whitney U test was used to compare the meibomian gland loss score, the blepharolipin score, and the number of Demodex in the two groups. The chi-square test was used to compare the positive detection rate of granular Demodex and the recurrence rate of the disease, P<0.05 was considered signi cant differences statistically.

Compare the Demodex infection
The detection rate of Demodex was 83.3% in patients with recurrent corneal epithelial erosion and 38.7% in the control group, with a signi cant difference between the 2 groups (χ²=7.60, P=0.00) ( Table 1). In the recurrent corneal erosion group, the maximum number of Demodex mites in one eye was 18, while in the control group, the maximum number of Demodex mites in one eye were 8.

Therapeutic Effect
There were 16 patients (16 eyes) including 9 males and 7 females in the treatment group, and 14 patients (14 eyes) including 8 males and 6 females in the observation group; there was no signi cant difference in age and gender the two groups. In the treatment group, after 3 times of IPL combined with meibomian gland massage treatments, the posterior eyelid congestion was reduced, the attachment of the secretion from the eyelash root was signi cantly reduced, the meibomian gland loss score and the blepharolipin score were all lower than before, and the number of Demodex and the detection rate of Demodex were signi cantly lower than before (Table 2). There was no signi cant change in the observation group.

Discussion
In 1994, Hope-Ross MW [4] has researched 30 patients with refractory recurrent corneal epithelial erosion with poor therapeutic effect, and he has found that all the patients had MGD, most of which have meibomian gland atrophy and deletion; he suggested that there was a signi cant correlation between MGD and RCES. In 2014, scholars studied 100 RCES patients (117 eyes), which revealing that 60 percent of the RCES eyes were with moderate or severe MGD simultaneously that need to be treated. The most susceptible corneal erosion area is the most susceptible corneal epithelium area that the meibomian gland and dry eye are affected. It was believed that the levels of matrix metalloproteinase (MMP) caused by meibomian gland dysfunction were involved in the occurrence of RCES [8] . In our research, it was also found that all patients of RCES had different degrees of meibomian gland dysfunction, the blepharolipin quality was lower than that of the control group, and the blepharolipin score was signi cantly higher than that of the control group; it was presented that meibomian gland dysfunction may be involved in the pathogenesis of RCES. This is consistent with the previous reports.
MGD was caused by various factors such as infection, in ammation, neurosecretory disorders, congenital abnormalities, and stellwag, while blepharitis would lead to severe MGD. It has been reported within China and overseas that there is a close relationship between the Demodex and anterior blepharitis; the Demodex was positively correlated with the symptom and evaluated scores; control the Demodex can signi cantly improve the symptoms of blepharitis [9.10] .
Studies have suggested that the main cause of meibomian gland obstruction is hyperkeratosis of the terminal duct [11,12] and opening. Meanwhile, the Demodex can be the bacteria carrier, bringing the bacteria to the palpebral edge during the process of secreting towards the palpebral edge. Plus, the decomposed esterase produced by bacteria may decompose the meibum into free fatty acids and other toxic medium, which will induce subclinical in ammation and release the pro-in ammatory factors, leading to increases the meibum viscosity or promotes epithelial keratinization of the glands, and cause the meibomian gland obstruction [13] . Therefore, Demodex also plays a vital role in the pathogenesis of MGD. We also found that almost all the RCES patients have angiotelectasis of the palpebral edge and hyperkeratosis of the meibomian gland opening; multiple forms of secretions can be found attached in the root of most patients' eyelash. All of the above are the characteristics of Demodex-infected blepharitis [14] , the detection rate of palpebral edge Demodex in these patients was much higher than that of normal patients. In our study, the detection rate of Demodex in the normal control group was only 38.1%, which was similar in previous literature [15] , while the detection rate of Demodex in RCES patients was 85%; this indicated that Demodex infection was very common in RCES patients. Chen Di [7] found in their study that the positive rate of Demodex in MGD patients was 86.4%. This also implied that a Demodex infection of the palpebral edge is involved in the pathogenesis of MGD and may be directly or indirectly involved in RCES patients' pathogenesis.
The current treatment of RCES mainly includes medication, wear corneal contact lens, mechanical and laser puncture of the anterior corneal stroma, and phototherapeutic keratectomy (PTK), etc. The main goal of the treatment is to relieve symptoms and reduce recurrence. Nevertheless, no treatment can prevent recurrence by now, and data shows that all treatments' recurrence rate was still above 20 percent [16][17][18] .
Treatment for MGD aims to ameliorate the HBCI of meibomian glands, improve the tear lm stability, and relieve the patients' discomfort symptoms. The current treatment includes physiotherapy such as clean palpebral edge, hot compress, meibomian glands massage, etc.; medication like arti cial tears eye drops, anti-in ammatory medicine, and antibiotic eye drops; along with diet therapy. IPL refers to a mature technology of dermatology for the treatment of skin telangiectasia, erythema, pigmentation, skin aging, and other diseases, and widely used for its good therapeutic effect. Toyos [19] et al. found that IPL improved patients' ocular surface with facial acne, and they proposed IPL as a potential treatment for MGD. Previous studies have indicated that IPL can be absorbed selectively by melanin and heme in deep skin tissues, causing the destruction and decomposition of pigment groups, the coagulation of blood vessels, and the elimination of abnormal telangiectasia, reducing vascularization of the palpebral edge of MGD patients [20] . In our study, after 3 times of IPL combined with a meibomian gland massage treatment, we found that the congestion of the palpebral edge was reduced, the secretion of meibum was signi cantly improved, and the deletion of the meibomian gland was also improved, which was similar to the results of other authors [21] . We con rmed that IPL could kill bacteria and Demodex [22] . Although Demodex still exist after three times of IPL treatment, the number of Demodex has decreased than before. Conversely, it is still unknown whether the palpebral edge's recovery is related to decreasing the Demodex number.
In our study, the ultimate goal of the treatment of MGD was to cure RCES. Surprisingly, we discovered that after 3 times treatments of IPL with meibomian gland massage therapy, the meibomian gland function and the amount of Demodex were signi cantly decreased, and corneal erosion of the patients has been effectively controlled. Plus, throughout the follow-up period, there was just 1 case that relapsed again. However, the recurrence rate was signi cantly less than before; the patients have only the corneal epithelium's punctate turbidity, compared to the recurrence before when the patients performed a large corneal epithelial defect, the improvement was remarkable. Our treatment was more effective than drug-only treatment. Our results showed that the treatment of MGD and Demodex is capable of effectively controlling RCES, which may offer a new approach to non-invasive treatment for the clinical treatment of RCES.
It is generally acknowledged that after 3 times of IPL treatment, the therapeutic effect can last 6 months to 1 year. Our study's follow-up period was 1 year; during this period, the disease was effectively controlled. Yet, our next study will be focused on whether there will be relapse as the follow-up period extends.
Meanwhile, on the therapeutic mechanism of RCES, whether treatment of IPL is only to improve the meibomian gland function and reduce the amount of Demodex is the direction of our future research.

Declarations
Ethics approval and consent to participate This retrospective study was performed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Wuhan Aier Eye Hanyang Hospital (HYEYE2018IRB01). All participants provided written informed consent.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated and analysed during the current study are not publicly available due trade secrets but are available from the corresponding author on reasonable request.

Figure 1
After treatment, the part of the corneal, which the epithelium erosion repeatedly showed punctuate opaque (white arrow), and not detached.