**4. Discussion**

The novel results of this strictly randomized, double-blinded, placebo-controlled clinical trial regarding the use of ultrasound-atomized *Houttuynia* eye drops for the treatment of MGD-related DED demonstrate that both the symptoms and signs of this condition are statistically significantly ameliorated from the baseline before treatment. Primary efficacy indicators were significantly enhanced compared with the placebo control group, and differences between the two groups were statistically significant. The specific clinical data were as follows: patients' eye symptom scores decreased by an average of 9 points, meibum quality decreased by nearly 1 point (indicating an improvement of nearly a grade), and TBUT increased to an average of 6.3 s after 4 weeks of treatment. The results of our subgroup analysis also show that the more severe the symptoms and the more turbid the nature of meibomian gland secretions, the better the therapeutic effect of *Houttuynia* eye drops will be. These results suggested that ultrasonically atomized *Houttuynia* eye drops had a clear positive effect on the treatment of MGD-related DED. This therapy can obviously relieve eye-related symptoms, improve meibum quality, and prolong tear break-up time in the clinic.

Inflammation is the main pathological manifestation of DED. This means that anti-inflammatory treatment is the most commonly used therapy for this condition. Inhibiting inflammation can stabilize the tear film and improve the clinical symptoms and signs of DED; indeed, the ultrasonic atomization of *Houttuynia* eye drops can improve both the symptoms and signs of DED, perhaps mainly related to the anti-inflammatory and antibacterial effects of this plant [11–13]. MGD-related DED is characterized by terminal duct obstruction and/or qualitative/quantitative changes in glandular secretions. This is because, on one hand, lipid discharge into the tear film is reduced, causing excessive tear evaporation and tear hyperosmolarity. Tear hyperosmolarity, therefore, also stimulates a cascade in the epithelial cells of the ocular surface, involving MAPK and NF-κB signaling pathways as well as the generation of inflammatory cytokines (i.e., interleukin-1 (IL-1α; IL-1β)), tumor necrosis factor-α (TNF-α)), and proteases. These activate and recruit inflammatory cells, induce inflammatory cascade reactions, and promote inflammation of the meibomian gland, eyelid margin, and ocular surface [25]. On the other hand, meibum stasis inside the gland can promote bacterial proliferation (including *Staphylococcus aureus* and other species); growth of these microorganisms enhances the production of lipid-degrading lipases and esterases that decompose the meibum into toxic mediators including free fatty acids. These toxic mediators increase meibum viscosity and melting temperature and result in tear film instability. Inflammation and the release of inflammatory cytokines can also occur; these phenomena aggravate the destruction of meibomian glands and, again, lead to a vicious circle [7]. The plant *H. cordata* is a traditional Chinese medicine therapy and has been shown to have significant anti-inflammatory and antibacterial effects. Indeed, results show that extracts from this species have anti-inflammatory effects on various cell and animal models. The main anti-inflammatory mechanism in this case acts to reduce the level of inflammation-related cytokines (i.e., TNF-α and IL-1β) and chemokines by inhibiting the NF-κB/MAPK pathway. It is clear that the NF-κB/MAPK pathway is one transduction inflammation pathway caused by MGD-related DED. One in vitro bacteriostatic test

showed that *H. cordata* exerts an obvious inhibitory e ffect on catarrhal bacteria, *S. aureus*, influenza bacillus, and pneumococcus [11,26]. Clinical observations have also shown that this treatment can inhibit bacteria on the ocular surface and can therefore be useful for bacterial conjunctivitis therapy. *Houttuynia* eye drops can inhibit bacterial growth on ocular surfaces and meibomian glands, thereby improving meibum quality.

The results of this study show that patients in the placebo group exhibited improved symptoms as well as all signs after four weeks of treatment. These data were statistically di fferent from those seen before treatment and are mainly related to the ultrasonic atomization treatment method. A number of clinical studies have shown that this method using di fferent liquid can improve meibum expressibility, stabilize the tear film, increase fluid, and mitigate the symptoms and signs of DED [10–12]. The placebo in this study was a sodium chloride solution. A previous study showed that eye atomization treatment with saline alone can also improve the symptoms and signs (including TBUT, corneal fluorescent staining, and Schirmer I test) of dry eye patients. Compared with the artificial tear (0.1% sodium hyaluronate eye fluid of Shentian Pharmaceutical Co., LTD., Shanghai, China) group, the therapeutic effect of the saline atomization group was significantly better (*p* < 0.05) [12]. Ultrasonic atomization makes droplets uniformly, continuously, and comprehensively act on the cornea, conjunctiva, and eyelid, maximizes the contact area between an eye and liquid, and therefore speeds up drug absorption [10]. Therefore, in this study, there was no statistical di fference in the change values of TBUT and tear secretion after 4 weeks of treatment, which may be related to the significant improvement e ffect of ultrasonic atomization on DED. This study has further clarified the e ffectiveness of this treatment.

Although the experimental group did not exhibit any statistical di fferences in terms of secondary efficacy indicators in comparison with the placebo group, data on the improvement of symptoms and signs in this former group were better than those in the latter. In addition, post hoc subgrouping analysis was conducted. Due to the limited number of samples, though the subgroup analysis cannot draw accurate conclusions, it can provide potential directions and ideas for the study. Thus, subsequent to grouping secondary indicators according to baseline severity, the data in this study show a statistical di fference between the two groups in terms of meibum expressibility and eyelid margin change. This may sugges<sup>t</sup> that patients with more severe meibomian gland blockage and worse eyelid margin state will experience more marked improvements as a result of *Houttuynia* eye drops when compared to the placebo set. This might be due to the therapeutic e ffects of the ultrasonic atomization method. No statistical di fferences were seen between conjunctival congestion and corneal fluorescent staining; this may be related to the low occurrence of severe conjunctival congestion and corneal fluorescent staining in the included patients. A further possibility might be that treatment times were not long enough and sample sizes were not big enough; larger samples and longer-term research projects will therefore be needed.

The results of visual acuity, IOP, and laboratory examinations after four weeks of treatment were basically unchanged and there were no adverse events related to the test drug. This shows again that this treatment is safe. Indeed, regarding subjective evaluation results, patients tolerated *Houttuynia* eye drop atomization treatments well, there was little irritation, and individuals were willing to tolerate this approach.

The main limitation of this study was that *Houttuynia* eye drops were not compared with other anti-inflammatory drugs. It is therefore unclear whether, or not, the therapeutic e ffect of *Houttuynia* eye drops is superior to other anti-inflammatory drugs commonly used in clinical practice. Inflammatory cytokines of tears were not detected during the delivery of this medication, another important indicator that can be used to evaluate eye surface inflammatory state. This is another important indicator that can be used to reflect *Houttuynia* eye drop mechanisms. Another major limitation was that there was missing correction for multiple comparisons. This study tested for multiple hypotheses. The test was considered valid when all three primary indicators were statistically significant. The three indices are independent of each other, and the results of the individual tests are essential, hence we made the decision not to adjust *p* values.

We have shown that treating patients with mild to moderate MGD-related DED for one month via the ultrasonic atomization of *Houttuynia* eye drops can significantly ameliorate the symptoms of eye discomfort, improve meibomian gland function, and stabilize tear films. The results of this study show that this treatment is safe.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2077-0383/9/12/4022/s1, Figure S1: The ultrasonic nebulizer, Table S1: Specific distribution of cases at each center, Table S2: Ocular symptom rating scale, Table S3: Characteristics of the patients at baseline, Table S4: Changes in individual eye symptom after 4-week treatment, Table S5: Correlation between indicator change and that at baseline, Table S6: Adverse events during the trial.

**Author Contributions:** Research design: Z.L. (Zhaolin Liu), M.J., and Z.L. (Zuguo Liu). Data acquisition and/or research execution: Z.L. (Zhaolin Liu), M.J., Y.L., J.L., X.X. (Xianghua Xiao), H.B., Z.P., H.S., X.X. (Xiaofeng Xie), M.Z., X.G., L.L., W.O., L.T., J.W., Y.Y., J.H., and Z.L. (Zuguo Liu). Data analysis: Z.L. (Zhaolin Liu), X.G., and L.L. Data interpretation and writing of the manuscript: Z.L. (Zhaolin Liu) and Z.L. (Zuguo Liu). All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported in part by the grants from China Association of Traditional Chinese Medicine and National Key R&D Program of China (grant number 2018YFA0107304).

**Acknowledgments:** The authors thank Yueping Zhou from Eye Institute of Xiamen University for his helpful suggestions and guidance.

**Conflicts of Interest:** The authors report no conflicts of interest in this work. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
