Okra eyelid patch versus sodium hyaluronate combined with ofloxacin eye drop in the treatment of meibomian gland dysfunction: a randomized controlled trial Authors: Yiting Wu, Xiaoping Ma, Min Li, Jiayi Ma Published: 2026 ABSTRACT BackgroundTo evaluate the clinical efficacy of okra eyelid patch versus sodium hyaluronate combined with ofloxacin eye drop in the treatment of patients suffering from meibomian gland dysfunction (MGD).Methods40 patients included were randomly divided into 2 groups: okra eyelid patch group (n = 20) and control group (n = 20). Before treatment, on the 14th and 28th day of treatment, the following ophthalmic examinations were performed: symptoms and signs score, Schirmer I test (SIT), tear film break-up time (TBUT), corneal fluorescein staining (CFS), optical quality analysis system (OQAS II) and ocular surface interferometer examination. Ocular surface disease index (OSDI) questionnaire score and in vivo confocal microscopy (IVCM) were performed before treatment and on the 28th day of treatment. Adverse effects were reported during the treatment period. Results were analyzed by SPSS 25.0 software. p < 0.05 was considered a statistically significant difference.ResultsOn day 28, okra eyelid patch group had a better curative effect in symptoms score, signs score, total score, OSDI questionnaire score, TBUT, and CFS score, lipid layer thickness and MASD compared with control group (p < 0.05). During the 28-day treatment, one patient in okra eyelid patch group reported eyelid redness, while the difference between the two groups was not statistically significant (p > 0.05).ConclusionsOkra eyelid patch has a more significant effect on MGD, especially in improving symptoms and signs, increasing lipid layer thickness, improving tear film stability, maintaining corneal epithelial integrity, and improving the quality of meibomian gland acini compared with artificial tears combining ofloxacin eye drops. It has few side effects, and provides a new therapeutic option for MGD associated with dry eye. Introduction Dry eye is a multifactorial chronic ocular surface disease characterized by tear film instability and ocular discomfort, and its etiology includes tear film instability, high osmotic pressure, ocular surface inflammation and injury, as well as neurosensory abnormalities caused by qualitative, quantitative, and kinetic changes of tear film. Basing on predominant etiology, dry eye is classified into two predominant and non-mutually exclusive categories; aqueous deficient dry eye (ADDE) and evaporative dry eye (EDE) [1]. In “China Dry Eye Expert Consensus: Definition and Classification (2020)” [2], experts gave a broader definition of dry eye and emphasized that dry eye can present with various types of discomfort and/or visual function impairment. In addition to revising the definition of dry eye, this consensus also developed three classifications of dry eye. According to the causes and risk factors, it can be classified into 7 different categories: systemic [3], ocular localization [4, 5], environmental [6, 7], lifestyle-related [8–10], surgical [11], medication [12–15], and others [16, 17]. Tear film is composed of lipid layer, aqueous layer, and mucin layer from the outside to the inside, and finally tears are evenly distributed on the ocular surface through tear dynamics (e.g., blinking, etc.). According to the abnormal composition or function of tears, dry eye is classified as aqueous tear deficiency dry eye, lipid deficiency dry eye, mucin deficiency dry eye, abnormal tear dynamics dry eye and mixed dry eye. According to the severity of dry eye signs, dry eye can also be classified into 3 categories: mild, moderate and severe. Meibomian gland dysfunction (MGD) refers to a chronic, diffuse meibomian gland pathology characterized by obstruction of the terminal ducts of the meibomian glands and/or qualitatively or quantitatively abnormal lipid secretion, which, clinically, results in abnormalities of tear film and inflammatory reactions on the ocular surface, leading to symptoms of eye irritation. If severe, it can also damage the cornea and then affect visual function [4]. Lipid deficiency dry eye accounts for the largest proportion of dry eye, and MGD is the leading cause of it, and its prevalence is increasing year by year. The results of a large international epidemiologic study from 11 countries [18] showed that out of 6525 patients, 54.3% were diagnosed with MGD, 61.8% were diagnosed with dry eye, and 67.6% had both dry eye and MGD. Traditional therapies for MGD [19] include hot compresses, spot cleaning, massage, topical or systemic application of anti-inflammatory drugs, etc. However, for some patients, the above traditional methods are not effective in relieving their symptoms. Recently, there are some new treatments such as intense pulsed light (IPL) and Lipiflow meibomian gland thermal pulsation therapy device. However, their high cost, need for in-clinic administration, and, in some cases, limited high-level evidence base may affect long-term patient accessibility and adherence for a chronic condition like MGD. This highlights the need for effective, convenient, and cost-effective therapeutic alternatives. Blepharitis is the main cause of MGD, and the lesions affect the eyelid skin, eyelash follicles, and meibomian glands [20]. Demodex mites (Arachnida, order Acaridae) are microscopic, elongated mites. Only two species are parasitic on human body, namely Demodex folliculorum and Demodex sebum. Eyelash follicles are the habitat of Demodex folliculorum, while meibomian glands and sebaceous glands are the habitat of Demodex sebum [21, 22]. Factors such as high oil secretion and secondary immune deficiencies (e.g., diabetes mellitus, malignant tumors, etc.) [23] will increase the possibility of Demodex infection. Demodex infection gradually progresses to mite blepharitis, which manifests as chronic inflammation of the eyelids and l meibomian glands, ultimately leading to MGD. Tea tree oil (TTO) can treat Demodex infection and relieve ocular surface inflammation related to Demodex mites [24]. It is currently recognized as a more effective treatment for c Demodex infection. Although TTO has a certain therapeutic effect on Demodex blepharitis, high concentrations or prolonged exposure to TTO and its pungent odor may cause mild to moderate ocular irritation and discomfort, especially in the elderly, children, pregnant women, or people with allergies [25]. A study comparing the anti-mite effects of tea tree essential oil eye patches with okra eye patches on palpebral Demodex mites found that okra eye patches and tea tree essential oil eye patches were comparable in mite removal and okra eye patches were less irritating than tea tree essential oil eye patches, which may be more suitable for sensitive elderly people, children, pregnant women, or allergy sufferers [26]. Okra is a widely known medicinal plant and has recently been found to be also rich in bioactive substances with antibacterial, anti-inflammatory and immunomodulatory properties. Okra has a high tannin content, which inhibits several common Gram-positive and Gram-negative bacteria [27]. Okra is also an immunomodulator with immunostimulatory and immunosuppressive activities [28, 29], which functions by promoting phagocyte activation and thereby increasing the levels of pro-inflammatory factors (e.g., TNF-α and IL-17) to improve host defenses against a variety of bacteria. Okra is rich in polysaccharides, which can effectively inhibit the growth of Demodex mites [26] and thus reduce the inflammatory response. The possible anti- Demodex mite mechanism of polysaccharides is to modulate the host’s immune system through activation of immune cells (e.g., lymphocytes, macrophages, and natural killer cells) [30]. In addition, okra is rich in flavonoids, polyphenols and vitamin C, which may act as a low-toxicity antioxidant and anti-inflammatory agent with few side effects [31, 32]. The reactive oxygen species (ROS) oxidation pathway has been confirmed to be involved in the pathogenesis of dry eye [33, 34], and okra may alleviate dry eye by inhibiting the ROS oxidation pathway to relieve ocular discomfort. In this study, okra eye patches were used to treat dry eye related to meibomian gland dysfunction, compared with sodium hyaluronate combined with ofloxacin eye drops, to investigate whether the effect is better and to provide a new treatment for meibomian gland dysfunction associated with dry eye. Materials and methods Subjects The study subjects were from 40 patients with MGD-associated dry eye who visited the Ophthalmology Clinic of Zhongshan Hospital Affiliated to Fudan University from July 2022 to February 2023. This clinical study was reviewed and approved by the Medical Ethics Committee of Zhongshan Hospital, Fudan University(Approval No.:B2022-441 R). The study was conducted in accordance with the ethical principles of the Declaration of Helsinki, and all the study subjects were aware of the nature of the study and the risks of the corresponding eye examinations, and written informed consents were obtained. Trial registration:ChiCTR2400091496(Retrospectively registered, Oct. 29th, 2024). Inclusion criteria Patients diagnosed with MGD associated with dry eye admitted to the Ophthalmology Clinic of Zhongshan Hospital, Fudan University. The diagnostic criteria for MGDwere specified as follows Abnormal eyelid margin and meibomian gland openings;Abnormal meibum secretion;Having eye symptoms;Meibomian gland loss;Abnormal thickness of lipid layer; Abnormal eyelid margin and meibomian gland openings; Abnormal eyelid margin and meibomian gland openings; Abnormal meibum secretion; Abnormal meibum secretion; Having eye symptoms; Having eye symptoms; Meibomian gland loss; Meibomian gland loss; Abnormal thickness of lipid layer; Abnormal thickness of lipid layer; Meibomian gland abnormalities can be diagnosed if any one of items (1) and (2) appears. Combined with item (3), those with eye symptoms are diagnosed as MGD, and those without eye symptoms are diagnosed as meibomian glands. abnormal function; Items (4) and (5) are indicators for strengthening diagnosis. If they appear alone, they have no diagnostic significance; Having symptoms or signs of dry eye at the same time, specifically as follows [35] Having one of the subjective symptoms such as dryness, discomfort, foreign body sensation, fatigue, burning sensation, redness of the eyes, fluctuation in vision, and an OSDI questionnaire score of ≥ 13 points;Tear break-up time (TBUT) ≤5 s or noninvasive break-up time (NIBUT) <10 s or Schirmer I test ≤ 5 mm/5 min; Having one of the subjective symptoms such as dryness, discomfort, foreign body sensation, fatigue, burning sensation, redness of the eyes, fluctuation in vision, and an OSDI questionnaire score of ≥ 13 points; Having one of the subjective symptoms such as dryness, discomfort, foreign body sensation, fatigue, burning sensation, redness of the eyes, fluctuation in vision, and an OSDI questionnaire score of ≥ 13 points; Tear break-up time (TBUT) ≤5 s or noninvasive break-up time (NIBUT) <10 s or Schirmer I test ≤ 5 mm/5 min; Tear break-up time (TBUT) ≤5 s or noninvasive break-up time (NIBUT) <10 s or Schirmer I test ≤ 5 mm/5 min; Dry eye is diagnosed when both (1) and (2) are met; Age: 18-85 years old; No gender limit; Exclusion criteria Suffering from glaucoma, tear duct obstruction, keratitis, iritis and other eye diseases and a history of eye trauma;History of corneal contact lens wear in the last 1 month;Lipiflow meibomian gland heat pulsation treatment, intense pulsed light therapy, and meibomian gland probing were performed in the last 1 month;History of eye surgery like cataract surgery, refractive surgery in the last 3 months;Systemic diseases such as severe skin diseases, systemic lupus erythematosus and dry syndrome;Allergy to okra ingredients;Using systemic antibiotics, glucocorticoids or immunosuppressants;Pregnancy or lactation. Suffering from glaucoma, tear duct obstruction, keratitis, iritis and other eye diseases and a history of eye trauma; Suffering from glaucoma, tear duct obstruction, keratitis, iritis and other eye diseases and a history of eye trauma; History of corneal contact lens wear in the last 1 month; History of corneal contact lens wear in the last 1 month; Lipiflow meibomian gland heat pulsation treatment, intense pulsed light therapy, and meibomian gland probing were performed in the last 1 month; Lipiflow meibomian gland heat pulsation treatment, intense pulsed light therapy, and meibomian gland probing were performed in the last 1 month; History of eye surgery like cataract surgery, refractive surgery in the last 3 months; History of eye surgery like cataract surgery, refractive surgery in the last 3 months; Systemic diseases such as severe skin diseases, systemic lupus erythematosus and dry syndrome; Systemic diseases such as severe skin diseases, systemic lupus erythematosus and dry syndrome; Allergy to okra ingredients; Allergy to okra ingredients; Using systemic antibiotics, glucocorticoids or immunosuppressants; Using systemic antibiotics, glucocorticoids or immunosuppressants; Pregnancy or lactation. Pregnancy or lactation. Methods Randomization and masking This was a prospective, randomized, open-label, controlled study. Forty study subjects were randomly divided into two groups of 20 each by means of a random number table (generated by SPSS 25.0). Due to the apparent physical differences between the intervention, the study was conducted in an open-label manner. Therefore, participants and clinicians were not masked to the treatment allocation. To ensure the objectivity of the results, the study relied heavily on instrument-based objective metrics. Furthermore, the analysis of key imaging data was performed by an assessor who was blinded to the group assignment. All clinical examinations followed standardized protocols. Treatment Okra Eye Patch Group: Participants applied a sterile okra eye patch (Yuejia, Shanghai Yuejia Industrial Co., Ltd.) nightly. The patch was placed on the closed eyelids for 10 minutes. After removal, the patch was used to gently wipe and cleanse the eyelash roots and meibomian gland orifices. Control Group: Participants administered ofloxacin eye drops (Santen Pharmaceutical Co., Ltd.) combined with 0.3% sodium hygroscopicity eye drops (S Santen Pharmaceutical Co., Ltd.) three times daily. A 10-minute interval was maintained between the instillation of the two drops. Common Procedure for Both Group: All participants received meibomian gland massage once every 2 weeks. The total treatment duration for both groups was 28 days (1 month). Outcomes Some basic information was gathered (name, age, date of birth, gender, etc.) and the examination was referred to eye findings. The following examinations were performed on patients before treatment respectively: symptom and sign scores, OSDI questionnaire scores, Schirmer I test (SIT), tear film break-up time (TBUT), corneal fluorescein staining (CFS), ocular surface interferometry to measure the lipid layer thickness, blink analysis, height of the tear river, and upper and lower meibomian gland loss areas, OQAS II system to measure the Tear Film Objective scatter index(TF-OSI), and in vivo laser confocal corneal microscopy to observe the change in the diameter of meibomian gland acini and the detection rate of Demodex mites. The primary outcome measures were symptom and sign scores, and the rest were secondary outcome measures. Follow-up visits were conducted on day 14 and day 28 respectively, and the above eye examinations were repeated at the follow-up visits. All pre- and post-treatment evaluations and examinations were performed by one physician. OSDI questionnaire was repeated only at the day 28 follow-up visit. In vivo laser confocal corneal microscopy was also repeated only at the 28th day follow-up as it was invasive. Both eyes of patients were examined and treated, but data from the right eye only were included in the analysis. During the treatment period, adverse reactions reported by the patients were promptly recorded, including eye redness, eye pain, foreign body sensation and eyelid edema. Serious adverse reactions may include blurred vision and acute eye infections. Symptoms and signs score Each patient was evaluated using a composite score encompassing five symptoms(itching, foreign body sensation, dryness, burning sensation and eyelid swelling) and seven clinical signs (meibomian gland secretion, number of plugged gland orifices, conjunctival injection, lid margin redness, lid margin debris, tear break up time (TBUT), and ocular surface staining with fluorescein), graded on a 4-point severity scale (0–3) [36] (Table 1). A standard slit lamp examination was performed to assess and record the severity of seven signs.Table 1Grading of five symptoms (1–5) and seven signs (6–12) in 100 patients with meibomian gland diseaseItemGrade 0Grade 1Grade 2Grade 3SymptomItchingForeign body sensationDrynessBurningEyelid swellingNoneNoneNoneNoneNoneAwarenessAwarenessAwarenessAwarenessNoticeableDesire to rubDesire to rubNeed dropsDesire to rubObviousFrequent rubDesire to close eyelidsFrequent dropsFrequent rubDecrease in palpebral fissureSignMG secretion (central lower eyelid)Plugged MG orifice (middle lower eyelid)Bulbar conjunctival rednessEyelid margin rednessEyelid margin debrisTear breakup time (second)Ocular surface stainingClearNoneNoneNoneNone>104-point categorical panels (Fig. 1)Cloudy<1/3PinkPink1–58–10Turbid with clumps1/3–2/3Light redLight red6–105–7Solid with paste>2/3Bright redBright red>10<5P: comparison between okra eye patch group and control group, OSDI: ocular surface disease index, SIT: Schirmer I test, TBUT: tear film break-up time, CFS: corneal fluorescein staining, TF-OSI: tear film objective scatter index, MALD: meibomian gland acinar longest diameter, MASD: meibomian gland acinar shortest diameter Table 1Grading of five symptoms (1–5) and seven signs (6–12) in 100 patients with meibomian gland diseaseItemGrade 0Grade 1Grade 2Grade 3SymptomItchingForeign body sensationDrynessBurningEyelid swellingNoneNoneNoneNoneNoneAwarenessAwarenessAwarenessAwarenessNoticeableDesire to rubDesire to rubNeed dropsDesire to rubObviousFrequent rubDesire to close eyelidsFrequent dropsFrequent rubDecrease in palpebral fissureSignMG secretion (central lower eyelid)Plugged MG orifice (middle lower eyelid)Bulbar conjunctival rednessEyelid margin rednessEyelid margin debrisTear breakup time (second)Ocular surface stainingClearNoneNoneNoneNone>104-point categorical panels (Fig. 1)Cloudy<1/3PinkPink1–58–10Turbid with clumps1/3–2/3Light redLight red6–105–7Solid with paste>2/3Bright redBright red>10<5P: comparison between okra eye patch group and control group, OSDI: ocular surface disease index, SIT: Schirmer I test, TBUT: tear film break-up time, CFS: corneal fluorescein staining, TF-OSI: tear film objective scatter index, MALD: meibomian gland acinar longest diameter, MASD: meibomian gland acinar shortest diameter Grading of five symptoms (1–5) and seven signs (6–12) in 100 patients with meibomian gland disease Symptom Itching Foreign body sensation Dryness Burning Eyelid swelling None None None None None Awareness Awareness Awareness Awareness Noticeable Desire to rub Desire to rub Need drops Desire to rub Obvious Frequent rub Desire to close eyelids Frequent drops Frequent rub Decrease in palpebral fissure Sign MG secretion (central lower eyelid) Plugged MG orifice (middle lower eyelid) Bulbar conjunctival redness Eyelid margin redness Eyelid margin debris Tear breakup time (second) Ocular surface staining Clear None None None None >10 4-point categorical panels (Fig. 1) Cloudy <1/3 Pink Pink 1–5 8–10 Turbid with clumps 1/3–2/3 Light red Light red 6–10 5–7 Solid with paste >2/3 Bright red Bright red >10 <5 P: comparison between okra eye patch group and control group, OSDI: ocular surface disease index, SIT: Schirmer I test, TBUT: tear film break-up time, CFS: corneal fluorescein staining, TF-OSI: tear film objective scatter index, MALD: meibomian gland acinar longest diameter, MASD: meibomian gland acinar shortest diameter Meibomian gland expression was performed (digital pressure) on the central third of the lower eyelid and graded on a 0–3 scale based on the worst type of secretion (clear, cloudy, turbid clumps, solid paste). Meibomian gland plugging was graded as follows: 0, clear orifice of meibomian glands in the central part of the lower eyelid; 1, less than one-third of the orifices contained turbid or oily secretion; 2, between one-third and two-thirds of the orifices contained turbid or oily secretion; 3, more than two-thirds of the orifices contained turbid or oily secretion. Bulbar conjunctival redness was graded as none, pink, light red, and bright red based on slit lamp examination. Lid margin redness was also graded as none, pink, light red, and bright red based on the color of the lower eyelid margin on slit lamp examination. Lid margin debris was scored based on the number of crusts at the lower eyelid margin, included in this study even though it is mostly present in anterior blepharitis. TBUT was performed before ocular surface staining. Fluorescein impregnated (single drop of saline) paper was placed in the lower fornix, and patient was asked to gently blink a few times and then stop blinking while being observed (blue light of slit lamp) for the first break on the corneal surface tear film. The time of the first split was recorded and graded on a 4-point categorical scale: grade 0 (>10 s), 1 (8–10s), 2 (5–7s), and 3 (<5 s). Ocular surface staining was immediately performed after recording the TBUT using a 4-point categorical scale panel (Fig. 1) [36, 37], which was a modification of panels introduced in the Oxford grading scale in order to be compatible with grading of the other signs. The patient was asked to look upward, downward, left, and right to expose the entire ocular surface. The panel which best represented the number and pattern of dots on the cornea and conjunctiva was selected and corresponding grade was recorded.Fig. 1Four-point categorical scale panel for grading ocular surface staining Four-point categorical scale panel for grading ocular surface staining OSDI questionnaire score All participants were administered the standardized OSDI questionnaire [38] to evaluate patients’ symptoms in the past week, including three dimensions as “eye symptoms”, “visual function” and “environmental triggers”, with a total of 12 questions. Points were scored according to the frequency of occurrence, and the final score was the ratio of the total score for all items to the total number of answered items multiplied by 25, with a total score of 0–100. A score of ≥ 13 was classified as dry eye, 13–22 as mild dry eye, 23–32 as moderate dry eye, and 33–100 as severe dry eye [35]. Schirmer I test(SIT) Schirmer I test (tear secretion test), which mainly reflects the secretion function (physiologic secretion) of the main lacrimal gland. Without using topical anesthesia, in a quiet darkroom environment with the same room temperature, the tear test filter paper strip (5 mm*35 mm, Tianjin Jingming) was folded back from the fold line, and placed into the middle and outer 1/3 of the conjunctival sac of the patient’s lower eyelid, and the patient was instructed to gently close eyes. 5 minutes later, the filter paper strip was gently taken out, and the length of the yellow area on the tear-soaked filter paper strip was read, and tear secretion value was recorded according to the length. The tear secretion value was recorded according to the length. Results were interpreted as follows: >10 mm (normal), 5–10 mm (low tear secretion), and <5 mm (insufficient tear secretion). Tear film break-up time(TBUT) When performing the tear film break-up time of the physical sign score, it was recorded. Corneal fluorescein staining(CFS) The examination method was the same as that of ocular surface fluorescein staining, and the corneal staining results were also evaluated during the ocular surface fluorescein staining examination. Using a 0-12point scale, the cornea was divided into 4 quadrants: superonasal, infranasal, superior temporal, and inferior temporal. For each quadrant, a score of 0–3 was assigned: 0 (no obvious staining), 1 (1–30 dot-like staining), 2 (>30 dot-like staining but no staining fusion), or 3 (punctate staining and fusion, filamentous staining or ulceration). Ocular surface interferometry An ocular surface interferometry [39] (Gaush iDea, Gaush Meditech Ltd) was utilized to measure the lipid layer thickness, blink analysis, height of the tear river, and upper and lower meibomian gland loss areas. Lipid layer thickness The device’s main lens barrel was inserted into the instrument, then the instrument was moved to focus on the lipid layer, and the video recording was initiated. During this period, the patient was asked to blink 3–5 times. After the recording was completed, the video was saved to obtain the lipid layer thickness value. Blink analysis After evaluating the lipid layer thickness, the same recorded video was automatically processed to assess blink quality and the blink quality (percentage) were generated. Height of the tear river The images of the patient’s eyes were acquired using the device’s placido lens tube (with concentric rings). The instrument was adjusted to focus on the lower tear meniscus, and the tear river height was measured from the captured images. Three repeated measurements were averaged to obtain the final result (Fig. 3). Height ≤ 0.2 mm was considered diagnostic as dry eye. Upper and lower meibomian gland loss areas The lens barrel was removed and imaging was conducted using infrared light only. The upper or lower eyelid was everted by a sterile cotton swab or a gloved finger to expose the tarsal conjunctiva and an image of the meibomian glands was then captured. After that, the glands were detected automatically and the missing area (%) of the meibomian glands was calculated and recorded, as shown in Fig. 2.Fig. 2Meibomian gland loss (the blue arrow points to the obvious meibomian gland loss area) Meibomian gland loss (the blue arrow points to the obvious meibomian gland loss area) OQAS II system The tear film objective scatter index (TF-OSI) was assessed using a dual-channel optical quality analysis system (OQAS II). Prior to the measurement, the patients’ refractive errors were completely corrected to remove the influence of the refractive error on the examination results. The patients were instructed to look at the internal visual target, and then the device was adjusted so that the patients’ eyes were centered and focused clearly on the screen. The Tear Film Analysis mode was selected to measure the tear film function, and patients were instructed refrain from blinking for 20 seconds during testing. After measurement, the tear film mean objective scatter index (Mean OSI) value was obtained, and the tear film objective scatter index (TF-OSI) value was calculated (TF-OSI = Mean OSI-OSI). Results evaluation criteria were interoreted as follows: TF-OSI < 0.6 indicated no dry eye, 0.6 ≤ TF-OSI < 1.2 indicated borderline dry eye, and TF-OSI ≥ 1.2 indicated dry eye (Fig. 3).Fig. 3Tear film optical function measurement. (A) TF-OSI < 0.6(0.11). (B) TF-OSI ≥ 1.2(1.49) Tear film optical function measurement. (A) TF-OSI < 0.6(0.11). (B) TF-OSI ≥ 1.2(1.49) In vivo confocal microscopy In vivo laser corneal confocal microscopy (IVCM) (Heidelberg, Germany) was used. The machine parameters were 670 nm laser wavelength, 400 µm × 400 µm observation field of view, 1 µm axial resolution. First, one drop of topical anesthetic (Proparacaine Hydrochloride Eye Drops, s.a. ALCON-COUVRUR n.v.) was instilled into the patient’s conjunctival sac. Subsequently, a sterile ophthalmic gel was applied into a disposable sterile contact cap, which was then placed in front of the examination lens. Patients were instructed to look straight up, with the forehead and chin securely fixed against at the rest respectively. The objective lens was carefully advanced until it made gentle contact with the center cornea and we observed the monitor to locate the optical interface between the tear film and corneal epithelial cells. Once this interface was identified, the focal plane was set to 0. Patients were asked to look upward, and the lower eyelid was everted to fully expose the meibomian glands. The objective lens were made to lightly contact the patient’s lower eyelid and the focal plane was adjusted to collect images of the meibomian gland structure near the eyelid margin. This study mainly observed patients’ meibomian gland acini and collected images of their structures. 1 high-definition MG image (3 images in total) was randomly selected from the nasal, temporal and central sides of each patient’s lower eyelid. Evaluation criteria: meibomian gland acinar longest diameter (MALD) and meibomian gland acinar shortest diameter (MASD), calculated using ImageJ software. In addition, the infection of Demodex in eyelash follicles and meibomian glands was also recorded. The presence of low-density rod-shaped shadows was considered as positive for Demodex. Adverse reactions During the treatment period, adverse reactions reported by the patients were promptly recorded, including eye redness, eye pain, foreign body sensation and eyelid edema. Serious adverse reactions may include blurred vision and acute eye infections. Statistical analysis The results were analyzed and processed using SPSS 25.0 software. The normality of all continuous data distributions was assessed using the Shapiro-West test. The measurement data, which conformed to a normal distribution, were expressed as mean ± standard deviation (x ± s). The chi-square test was used to compare gender and Demodex mite detection rates. Comparison of measurement data between two groups was performed using independent samples t test. Comparison of measurement data before and after treatment in the same group was performed using paired sample t test. Given the exploratory analysis of multiple secondary endpoints, no formal correction for multiple comparisons was applied; p-values for these comparisons should be interpreted with caution. When p < 0.05, the difference was statistically significant. ImageJ software was used to analyze and process the images of meibomian gland acini. Results Baseline information The subjects of this study were derived from 40 patients (40 eyes) who attended the outpatient ophthalmology clinic of Zhongshan Hospital, Fudan University, from July 2022 to February 2023. The baseline information is shown in Table 2.Table 2Baseline informationOkra eye patch group(n = 20)Control group(n = 20)PAge(years old,x ± s)39.55 ± 14.0239.55 ± 15.880.705Sex,n(%)Male9(45%)8(40%) 0.749FemaleSymptom score11(55%)6.75 ± 2.2212(60%)6.50 ± 1.730.694Sign score11.65 ± 1.5711.80 ± 1.990.792Total score18.40 ± 2.8518.30 ± 3.030.915OSDI questionnaire scoreSIT(mm/5 min)27.40 ± 8.3411.25 ± 8.3529.27 ± 6.7111.70 ± 8.500.4380.867TBUT(s)5.23 ± 1.845.05 ± 1.220.714CFSLipid layer thickness(mm)4.00 ± 2.3265.05 ± 18.663.75 ± 2.3171.95 ± 21.430.7350.284Blink analysis(%)Tear river height(mm)51.95 ± 22.130.13 ± 0.0555.70 ± 23.870.13 ± 0.080.6090.700Upper MG loss area(%)Lower MG loss area(%)TF-OSIMALD (µm)MASD (µm)Demodex mite detection rate(%)38.95 ± 7.2635.80 ± 17.100.89 ± 0.74115.05 ± 24.2839.30 ± 11.3920%38.75 ± 6.3332.15 ± 13.150.75 ± 0.95111.32 ± 22.8036.32 ± 10.7915%0.9270.4540.6030.6190.3990.677 Table 2Baseline informationOkra eye patch group(n = 20)Control group(n = 20)PAge(years old,x ± s)39.55 ± 14.0239.55 ± 15.880.705Sex,n(%)Male9(45%)8(40%) 0.749FemaleSymptom score11(55%)6.75 ± 2.2212(60%)6.50 ± 1.730.694Sign score11.65 ± 1.5711.80 ± 1.990.792Total score18.40 ± 2.8518.30 ± 3.030.915OSDI questionnaire scoreSIT(mm/5 min)27.40 ± 8.3411.25 ± 8.3529.27 ± 6.7111.70 ± 8.500.4380.867TBUT(s)5.23 ± 1.845.05 ± 1.220.714CFSLipid layer thickness(mm)4.00 ± 2.3265.05 ± 18.663.75 ± 2.3171.95 ± 21.430.7350.284Blink analysis(%)Tear river height(mm)51.95 ± 22.130.13 ± 0.0555.70 ± 23.870.13 ± 0.080.6090.700Upper MG loss area(%)Lower MG loss area(%)TF-OSIMALD (µm)MASD (µm)Demodex mite detection rate(%)38.95 ± 7.2635.80 ± 17.100.89 ± 0.74115.05 ± 24.2839.30 ± 11.3920%38.75 ± 6.3332.15 ± 13.150.75 ± 0.95111.32 ± 22.8036.32 ± 10.7915%0.9270.4540.6030.6190.3990.677 Baseline information Female Symptom score 11(55%) 6.75 ± 2.22 12(60%) 6.50 ± 1.73 OSDI questionnaire score SIT(mm/5 min) 27.40 ± 8.34 11.25 ± 8.35 29.27 ± 6.71 11.70 ± 8.50 0.438 0.867 CFS Lipid layer thickness(mm) 4.00 ± 2.32 65.05 ± 18.66 3.75 ± 2.31 71.95 ± 21.43 0.735 0.284 Blink analysis(%) Tear river height(mm) 51.95 ± 22.13 0.13 ± 0.05 55.70 ± 23.87 0.13 ± 0.08 0.609 0.700 Upper MG loss area(%) Lower MG loss area(%) TF-OSI MALD (µm) MASD (µm) Demodex mite detection rate(%) 38.95 ± 7.26 35.80 ± 17.10 0.89 ± 0.74 115.05 ± 24.28 39.30 ± 11.39 20% 38.75 ± 6.33 32.15 ± 13.15 0.75 ± 0.95 111.32 ± 22.80 36.32 ± 10.79 15% 0.927 0.454 0.603 0.619 0.399 0.677 There were 20 patients in okra eye patch group, of which 45% were males, totaling 9 patients, and 55% were females, totaling patients. The average age was 39.55 ± 14.02 years old, ranging from 26 to 74, with 12 (60%) aged 18–40, 6 (30%) aged 41–60, 2 (10%) aged 61–85. There were 20 patients in the control group, 8 males (40%) and 12 females (60%). The average age was 39.55 ± 15.88 years, ranging from 24 to 83, with 13 (65%) aged 18–40, 5 (25%) aged 41–60, and 2 (10%) aged 61–85. Before treatment, the symptom score of the okra eye patch group was 6.75 ± 2.22 points, the sign score was 11.65 ± 1.57 points, and the total score was 18.40 ± 2.85 points. The symptom score of the control group was 6.50 ± 1.73 points, the signs score was 11.80 ± 1.99 points, and the total score was 18.30 ± 3.03 points. Before treatment, the OSDI questionnaire score in the okra eye patch group was 27.40 ± 8.34 points, and that in the control group was 29.27 ± 6.71 points. Before treatment, Schirmer I test of the okra eye patch group was 11.25 ± 8.35 mm/5 min, and the control group was 11.70 ± 8.50 mm/5 min. TBUT of the okra eye patch group was 5.23 ± 1.84 s, and the control group was 5.05 ± 1.22 s. CFS of the okra eye patch group was 4.00 ± 2.32 points, and that of the control group was 3.75 ± 2.31 points. The ocular surface interferometer was detected before treatment. The lipid layer thickness of the okra eye patch group was 65.05 ± 18.66 mm, the blink analysis was 51.95 ± 22.13%, and the tear river height was 0.13 ± 0.05 mm. The upper and lower missing meibomian glands area was 38.95 ± 7.26% and 35.80 ± 17.10% respectively. As for the control group, the thickness of the lipid layer was 71.95 ± 21.43 mm, the blink analysis was 55.70 ± 23.87% and the height of the tear river was 0.13 ± 0.08 mm. The missing area of the upper and lower meibomian glands was 38.75 ± 6.33% and 32.15 ± 13.15%. The results of the OQAS II system before treatment showed that the TF-OSI of the okra eye patch group was 0.89 ± 0.74 and that of the control group was 0.75 ± 0.95. The IVCM results before treatment showed that the MALD of the okra eye patch group was 115.05 ± 24.28 µm, the MASD was 39.30 ± 11.39 µm and the detection rate of Demodex was 20%. While, the MALD of the control group was 111.32 ± 22.80 µm, the MASD was 36.32 ± 10.79 µm, and the detection rate of Demodex was 15%. Comparing the baseline data above between the two groups before treatment, the p values were all > 0.05, indicating comparability between the two groups. Comparison of efficacy before and after treatment in okra eye patch group All assessment results before treatment, on the 14th day, and on the 28th day of treatment are compared in Table 3.Table 3Comparisons before and after okra eye patch treatmentItemBeforeDay 14Day 28P1P2Symptom score6.75 ± 2.224.15 ± 2.182.55 ± 1.57<0.001<0.001Sign score11.65 ± 1.579.95 ± 2.436.45 ± 3.150.034<0.001Total score18.40 ± 2.8514.10 ± 3.399.00 ± 4.15<0.001<0.001OSDISIT(mm/5 min)TBUT(s)CFS27.40 ± 8.3411.25 ± 8.355.23 ± 1.844.00 ± 2.32-14.55 ± 8.166.42 ± 2.052.45 ± 2.0118.33 ± 6.7519.60 ± 8.009.80 ± 2.831.45 ± 1.54-0.2070.1060.0160.0010.002<0.001<0.001Lipid layer thickness65.05 ± 18.6665.30 ± 18.0886.20 ± 17.330.965<0.001Blink analysisTear river height51.95 ± 22.130.13 ± 0.0571.40 ± 29.440.14 ± 0.0481.20 ± 24.460.17 ± 0.040.0190.2220.0010.007Upper MG loss areaLower MG loss areaTF-OSIMALD(µm)MASD(µm)detection rate of Demodex38.95 ± 7.2635.80 ± 17.100.89 ± 0.74115.05 ± 24.2839.30 ± 11.3920%36.05 ± 5.3229.60 ± 11.160.60 ± 0.62---34.45 ± 6.3027.10 ± 10.480.44 ± 0.4794.32 ± 17.8628.60 ± 8.555%0.1540.1440.134---0.0290.0290.0230.0070.0030.139P1: comparison before treatment and day 14,P2: comparison before treatment and day 28 Table 3Comparisons before and after okra eye patch treatmentItemBeforeDay 14Day 28P1P2Symptom score6.75 ± 2.224.15 ± 2.182.55 ± 1.57<0.001<0.001Sign score11.65 ± 1.579.95 ± 2.436.45 ± 3.150.034<0.001Total score18.40 ± 2.8514.10 ± 3.399.00 ± 4.15<0.001<0.001OSDISIT(mm/5 min)TBUT(s)CFS27.40 ± 8.3411.25 ± 8.355.23 ± 1.844.00 ± 2.32-14.55 ± 8.166.42 ± 2.052.45 ± 2.0118.33 ± 6.7519.60 ± 8.009.80 ± 2.831.45 ± 1.54-0.2070.1060.0160.0010.002<0.001<0.001Lipid layer thickness65.05 ± 18.6665.30 ± 18.0886.20 ± 17.330.965<0.001Blink analysisTear river height51.95 ± 22.130.13 ± 0.0571.40 ± 29.440.14 ± 0.0481.20 ± 24.460.17 ± 0.040.0190.2220.0010.007Upper MG loss areaLower MG loss areaTF-OSIMALD(µm)MASD(µm)detection rate of Demodex38.95 ± 7.2635.80 ± 17.100.89 ± 0.74115.05 ± 24.2839.30 ± 11.3920%36.05 ± 5.3229.60 ± 11.160.60 ± 0.62---34.45 ± 6.3027.10 ± 10.480.44 ± 0.4794.32 ± 17.8628.60 ± 8.555%0.1540.1440.134---0.0290.0290.0230.0070.0030.139P1: comparison before treatment and day 14,P2: comparison before treatment and day 28 Comparisons before and after okra eye patch treatment OSDI SIT(mm/5 min) TBUT(s) CFS 27.40 ± 8.34 11.25 ± 8.35 5.23 ± 1.84 4.00 ± 2.32 - 14.55 ± 8.16 6.42 ± 2.05 2.45 ± 2.01 18.33 ± 6.75 19.60 ± 8.00 9.80 ± 2.83 1.45 ± 1.54 - 0.207 0.106 0.016 0.001 0.002 <0.001 <0.001 Blink analysis Tear river height 51.95 ± 22.13 0.13 ± 0.05 71.40 ± 29.44 0.14 ± 0.04 81.20 ± 24.46 0.17 ± 0.04 0.019 0.222 0.001 0.007 Upper MG loss area Lower MG loss area TF-OSI MALD(µm) MASD(µm) detection rate of Demodex 38.95 ± 7.26 35.80 ± 17.10 0.89 ± 0.74 115.05 ± 24.28 39.30 ± 11.39 20% 36.05 ± 5.32 29.60 ± 11.16 0.60 ± 0.62 - - - 34.45 ± 6.30 27.10 ± 10.48 0.44 ± 0.47 94.32 ± 17.86 28.60 ± 8.55 5% 0.154 0.144 0.134 - - - 0.029 0.029 0.023 0.007 0.003 0.139 P1: comparison before treatment and day 14,P2: comparison before treatment and day 28 After treated by okra eyelid patch, symptoms score, total score, CFS score, and blink analysis of patients on day 14 were significantly improved. The differences were statistically significant (p < 0.05). While signs score, SIT, TBUT, lipid layer thickness, tear meniscus height, the missing area of upper and lower meibomian gland, and TF-OSI had no statistical significance (p > 0.05). On day 28, symptoms score, signs score, total score, OSDI questionnaire score, SIT, TBUT, CFS score, lipid layer thickness, blink analysis, tear meniscus height, upper and lower meibomian gland missing area, TF-OSI, MALD and MASD in the group of okra eyelid patch were significantly improved compared with those before treatment, and the differences were statistically significance (p < 0.05). The detection rate of Demodex mites decreased after treating, but the difference was not statistically significant (p > 0.05). Comparison of anterior segment photos is shown in Fig. 4. The comparison of corneal fluorescein staining before treatment, on the 14th day, and on the 28th day is shown in Fig. 5. The morphology of Demodex mites under IVCM is shown in Fig. 6.Fig. 4Anterior segment photos of the okra eye patch group before and after treatment: (A) obvious bulbar conjunctival congestion; (B) congestive eyelid margin with debris (circled in blue); (C) blocked meibomian gland openings and turbid secretions (pointed out by the blue arrow); (D) Bulbar conjunctival congestion having been improved; (E) No obvious congestion in the eyelid margin; (F) unobstructed meibomian gland opening and clear secretion. Picture A, B, and C were before treatment, and pictures D, E, and F were after treatmentFig. 5Comparison of CFS before and after okra eye patch treatmentFig. 6Morphology of Demodex mites under IVCM (mid-low reflective shadow in stick shape pointed out by the blue arrow) Anterior segment photos of the okra eye patch group before and after treatment: (A) obvious bulbar conjunctival congestion; (B) congestive eyelid margin with debris (circled in blue); (C) blocked meibomian gland openings and turbid secretions (pointed out by the blue arrow); (D) Bulbar conjunctival congestion having been improved; (E) No obvious congestion in the eyelid margin; (F) unobstructed meibomian gland opening and clear secretion. Picture A, B, and C were before treatment, and pictures D, E, and F were after treatment Comparison of CFS before and after okra eye patch treatment Morphology of Demodex mites under IVCM (mid-low reflective shadow in stick shape pointed out by the blue arrow) Comparison of the curative effect after treatment between the okra eye patch group and the control group All assessment results between the two groups on day 14 and 28 are compared in Tables 4 and 5 respectively.Table 4Comparisons between the two groups on day 14ItemOkra eye patch groupControl groupPSymptom score4.15 ± 2.184.70 ± 1.560.365Sign score9.95 ± 2.4311.20 ± 2.400.110Total score14.10 ± 3.3915.90 ± 3.010.083SIT(mm/5 min)14.55 ± 8.1612.05 ± 7.980.334TBUT(s)6.42 ± 2.055.43 ± 1.290.078CFS2.45 ± 2.013.35 ± 1.930.157lipid layer thickness (mm)65.30 ± 18.0859.20 ± 20.620.326blink analysis (%)tear river height (mm)71.40 ± 29.440.14 ± 0.0469.60 ± 27.820.15 ± 0.090.8430.703upper MG loss area (%)lower MG loss area (%)36.05 ± 5.3229.60 ± 11.1637.70 ± 6.5833.21 ± 9.920.3880.293TF-OSI0.60 ± 0.620.67 ± 0.820.750P: comparison of the two groups on day 14Table 5Comparisons between the two groups on day 28ItemOkra eye patch groupControl groupPSymptom score2.55 ± 1.574.00 ± 1.300.003Sign score6.45 ± 3.159.50 ± 1.850.001Total scoreOSDISIT(mm/5 min)TBUT(s)CFS9.00 ± 4.1518.33 ± 6.7519.60 ± 8.009.80 ± 2.831.45 ± 1.5413.50 ± 2.6923.65 ± 5.6315.05 ± 8.236.48 ± 1.472.70 ± 1.45<0.0010.0100.084<0.0010.012lipid layer thickness (mm)86.20 ± 17.3370.10 ± 23.940.020blink analysis (%)tear river height (mm)81.20 ± 24.460.17 ± 0.0474.70 ± 24.130.15 ± 0.050.4050.231upper MG loss area (%)lower MG loss area (%)34.45 ± 6.3027.10 ± 10.4838.15 ± 5.8729.70 ± 11.660.0620.463TF-OSI0.44 ± 0.470.50 ± 0.870.776MALD(µm)94.32 ± 17.86100.08 ± 18.360.322MASD(µm)detection rate of Demodex28.65 ± 8.555%35.26 ± 8.9615%0.0220.282P:comparison of the two groups on day 28 Table 4Comparisons between the two groups on day 14ItemOkra eye patch groupControl groupPSymptom score4.15 ± 2.184.70 ± 1.560.365Sign score9.95 ± 2.4311.20 ± 2.400.110Total score14.10 ± 3.3915.90 ± 3.010.083SIT(mm/5 min)14.55 ± 8.1612.05 ± 7.980.334TBUT(s)6.42 ± 2.055.43 ± 1.290.078CFS2.45 ± 2.013.35 ± 1.930.157lipid layer thickness (mm)65.30 ± 18.0859.20 ± 20.620.326blink analysis (%)tear river height (mm)71.40 ± 29.440.14 ± 0.0469.60 ± 27.820.15 ± 0.090.8430.703upper MG loss area (%)lower MG loss area (%)36.05 ± 5.3229.60 ± 11.1637.70 ± 6.5833.21 ± 9.920.3880.293TF-OSI0.60 ± 0.620.67 ± 0.820.750P: comparison of the two groups on day 14 Comparisons between the two groups on day 14 blink analysis (%) tear river height (mm) 71.40 ± 29.44 0.14 ± 0.04 69.60 ± 27.82 0.15 ± 0.09 0.843 0.703 upper MG loss area (%) lower MG loss area (%) 36.05 ± 5.32 29.60 ± 11.16 37.70 ± 6.58 33.21 ± 9.92 0.388 0.293 P: comparison of the two groups on day 14 Table 5Comparisons between the two groups on day 28ItemOkra eye patch groupControl groupPSymptom score2.55 ± 1.574.00 ± 1.300.003Sign score6.45 ± 3.159.50 ± 1.850.001Total scoreOSDISIT(mm/5 min)TBUT(s)CFS9.00 ± 4.1518.33 ± 6.7519.60 ± 8.009.80 ± 2.831.45 ± 1.5413.50 ± 2.6923.65 ± 5.6315.05 ± 8.236.48 ± 1.472.70 ± 1.45<0.0010.0100.084<0.0010.012lipid layer thickness (mm)86.20 ± 17.3370.10 ± 23.940.020blink analysis (%)tear river height (mm)81.20 ± 24.460.17 ± 0.0474.70 ± 24.130.15 ± 0.050.4050.231upper MG loss area (%)lower MG loss area (%)34.45 ± 6.3027.10 ± 10.4838.15 ± 5.8729.70 ± 11.660.0620.463TF-OSI0.44 ± 0.470.50 ± 0.870.776MALD(µm)94.32 ± 17.86100.08 ± 18.360.322MASD(µm)detection rate of Demodex28.65 ± 8.555%35.26 ± 8.9615%0.0220.282P:comparison of the two groups on day 28 Comparisons between the two groups on day 28 Total score OSDI SIT(mm/5 min) TBUT(s) CFS 9.00 ± 4.15 18.33 ± 6.75 19.60 ± 8.00 9.80 ± 2.83 1.45 ± 1.54 13.50 ± 2.69 23.65 ± 5.63 15.05 ± 8.23 6.48 ± 1.47 2.70 ± 1.45 <0.001 0.010 0.084 <0.001 0.012 blink analysis (%) tear river height (mm) 81.20 ± 24.46 0.17 ± 0.04 74.70 ± 24.13 0.15 ± 0.05 0.405 0.231 upper MG loss area (%) lower MG loss area (%) 34.45 ± 6.30 27.10 ± 10.48 38.15 ± 5.87 29.70 ± 11.66 0.062 0.463 MASD(µm) detection rate of Demodex 28.65 ± 8.55 5% 35.26 ± 8.96 15% 0.022 0.282 P:comparison of the two groups on day 28 On the 14th day, the effects of two groups were similar, and the differences were not statistically significant (p > 0.05). On day 28, compared between the okra eyelid patch group and the control group, symptoms score, signs score, total score, OSDI questionnaire score, TBUT, and CFS score, lipid layer thickness and MASD had a better curative effect, and the differences were statistically significant (p < 0.05). While as for SIT, blink analysis, tear meniscus height, upper and lower meibomian gland missing area, TF-OSI, MALD and detection rate of demodex mites, the two had similar effects and the differences were not statistically significant (p > 0.05). Adverse reaction reports During the 28 days of treatment, the okra eye patch demonstrated a safety profile. A single, mild adverse event of transient eyelid skin redness was reported in the okra group (1/20,5%), which resolved completely within 24 hours after proper cleansing. Discussion This study demonstrates the a 28-day treatment with the okra eye patch significantly improved both symptoms and signs in patients with MGD-related dry eye and was more effective than conventional therapy in increasing tear film layer thickness. This study compares the therapeutic effects of okra eye patch and ofloxacin combined with sodium hyaluronate on MGD-related dry eye to explore whether okra eye patch can effectively relieve patient discomfort and provide new methods for clinical treatment. In terms of outcome evaluation, in addition to common meibomian gland and dry eye examinations such as symptom scores, signs scores, SIT, TBUT, and CFS, three instruments including IVCM, ocular surface interferometry, and OQAS II system were also used. IVCM mainly focuses on observing the morphological changes of meibomian gland acini and Demodex infection. The ocular surface interferometer mainly focuses on observing changes on the missing area of meibomian glands, lipid layer thickness and so on. The OQAS II system focuses on observing dynamic tear film function. MGD is mainly due to prevent smooth drainage of meibum and accumulate in the meibomian gland openings, resulting in pathological expansion of the acini of the meibomian gland opening machines. Under a confocal microscope, the meibomian gland acini became enlarged, the density decreased, the meibum in the cavity was highlighted, and the meibomian gland openings were blocked and expanded. Under the microscope, immune-related dry eye can show enlarged corneal epithelial cells, activated Langerhans cells in the subepithelial and palpebral conjunctiva, tortuous nerve fibers in the superficial stroma, increased inflammatory cells, and even meibomian gland fibrosis. We made the results more convincing by observing the changes in the meibomian gland acini combined with the improvement of the patients’ clinical symptoms. IVCM has rich observation indicators for meibomian glands, including acinar morphological changes (expansion and atrophy) of meibomian glands, infiltration status of inflammatory cells, tissue fibrosis grade, and Demodex infection in eyelash follicles and meibomian glands. Situation etc. According to the study by McCann et al. [40], compared with the meibomian glands of the upper eyelid, the meibomian glands of the lower eyelid are lost to a greater extent. Meibomian gland lesions are evaluated through changes in the meibomian gland acini of the lower eyelid, which are more sensitive than those in the upper eyelid. This is the reason why this study observed the morphological changes of the meibomian glands in the lower eyelid to evaluate the meibomian glands. In the past, IVCM was used to observe the characteristics of meibomian glands was mostly limited to three locations: nasal, central and temporal, and the average value was calculated to represent the parameters of the entire lower eyelid acini [41, 42]. Liang Qingfeng et al. [43] found that when observing the morphology of meibomian gland acini in MGD patients under IVCM, most of the acini near the eyelid margin showed irregular expansion, and a large number of high-altitude acini could be observed in the openings of the meibomian glands and in the acini. Reflective material is deposited, while most of the acini near the fornix are in a shrunken state, the cubic cell structure around them disappears, the surrounding tissue appears fibrotic, fishnet-like or cord-like, and the normal meibomian gland acinar structure disappears. The results of the study suggest that the structural morphology and pathological processes of meibomian gland acini are markedly different in distinct regions. Therefore, in this study, in order to observe changes in acinar diameter, the structure of meibomian gland acini near the eyelid margin was selected. One high-definition meibomian gland image (3 images in total) was selected from the nasal side, temporal side, and center of the lower eyelid margin respectively, and the average value was calculated as the parameter of the meibomian gland acini of the lower eyelid margin. While IVCM acquired images of the meibomian glands, it could quickly scan the meibomian glands and eyelash follicles to see if Demodex mites exist and accumulate at the same time. However, it could not accurately count, so this study used the detection rate to describe Demodex infection. After applying okra eye patch for 28 days, the patient’s MALD and MASD were significantly smaller than before treatment, and on the 28th day, the MASD results were significantly different between the okra eye patch group and the control group, indicating that okra eye patch can effectively improve the expansion of meibomian gland acini due to inflammatory infiltration in MGD patients. On the 28th day, we also found that the detection rate of Demodex in the okra eye patch group dropped from 20% before treatment to 5%, which is consistent with previous research results [26], indicating that the okra eye patch can effectively eliminate mites. There was no statistically significant difference in the detection rate of Demodex between the okra eye patch group and the control group after 28 days of treatment. The possible reason is that the number of participants in this study was limited, and the main research subjects are not patients with mite blepharitis. We only observe Demodex infection under IVCM. The ocular surface interferometer (Gaush iDea, Gaush Meditech Ltd) is a new type of dry eye diagnostic equipment for automatic non-invasive examination. It does not affect the quality and quantity of the tear film and has good sensitivity and specificity for clinical screening and diagnosis of dry eye [39]. It can measure various dry eye parameters such as lipid layer thickness, blink analysis, tear river height, and upper and lower meibomian gland loss areas. Among them, the thickness of the lipid layer and the missing area of the upper and lower meibomian glands are objective manifestations of the meibomian gland function. Quantitative detection, quantitative analysis and comparison can reflect the quality of the meibomian glands in patients with MGD-related dry eye. The results on the 28th day showed that compared with the control group, the okra eye patch group had a more significant improvement in lipid layer thickness, which may further improve the availability of meibum at the eyelid edge and tear film, and alleviate patients’ tear evaporation, tear film hypertonicity and instability. There was little difference between the two groups in the detection of meibomian gland loss area by ocular surface interferometry. Combined with the obvious improvement results of IVCM on MASD, although the diameter of the patient’s meibomian gland acini decreased after treatment, the treatment time is only 1 month, and the recoverable loss area of the meibomian glands may take longer to observe. The OQAS II system is an optical device that can objectively, quantitatively, comprehensively and accurately analyze visual quality. Its dual-channel technology can directly collect point light sources imaged on the retina, then the point spread function is analyzed to obtain the relevant parameters of visual quality, such as objective scatter index (OSI), etc. The patient’s tear film function is measured in the Tear Film Analysis mode. After the 20-second detection process is completed, the tear film average objective scattering index (Mean OSI) can be obtained. Then the tear film dynamic objective scattering index (TF-OSI) can be calculated (the average objective scattering index minus the basic objective scattering index), that is, TF-OSI = Mean OSI-OSI. The tear film is the only optical medium that changes dynamically. The OQAS II system can objectively and dynamically examine tear film function, assist in diagnosis and grading, and evaluate therapeutic efficacy [44]. The results on day 28 showed that there was no significant difference in TF-OSI between the okra eye patch group and the control group, and both treatments could improve the patient’s tear film function. This study ultimately found that okra eye patches are highly effective in treating MGD-related dry eye, especially in terms of increasing lipid layer thickness, improving tear film stability, maintaining corneal epithelial integrity, and improving meibomian gland acini quality, which is better than artificial tear supplementation and topical antibacterial drug application. This study explores the efficacy of okra eye patch in the treatment of meibomian gland dysfunction associated with dry eye from multiple perspectives including symptoms, signs, ocular surface interferometry, OQAS II system and IVCM, which is more objective and complete. When considering our findings in the context of available therapies, the okra eye patch represents a complementary approach with distinct practical advantages. While devices like intense pulsed light (IPL) [45] and Lipiflow thermal pulsation therapy [46] have demonstrated efficacy, their utility can be constrained by cost, the need for repeated clinic visits, and a relative scarcity of large-scale RCTs on long-term outcomes [47]. In contrast, the okra eye patch offers a low-cost and home-based intervention. In a study on the alternative treatment of TTO eye patches, it was found that okra eye patches are equivalent to TTO eye patches. They have an excellent performance in removing mites and suppressing mites. Besides, they are gentler and have better tolerance as well as wider applicability [26]. Our study also confirmed this point. Only one mild and self-limiting case of local skin redness was observed in the okra group, and importantly, the incidence of adverse reactions did not differ significantly from that in the control group(p = 0.235). While both interventions appeared safe in this short-term study, coupled with its comparable efficacy, supports its potential as a gentler alternative. This favorable safety profile is crucial for chronic conditions like MGD, which requires long-term management strategies that patients can adhere to comfortably. This study has certain limitations. First, the open-label design represents a limitation. Due to the distinct physical forms of the interventions (patch vs eye drops), participants and clinicians were not masked to treatment allocation. This may have introduced expectation bias, particularly for patient-reported outcomes (symptoms, OSDI). Although we prioritized objective instrumental endpoints and performed blinded image analysis, the lack of masking for subjective assessments remains a limitation. Future double-masked, sham-controlled trials are warranted. Second, the absence of a sham/vehicle-controlled eye patch group prevents determination of whether the observed benefits are specifically attributable to the okra components or result from non-specific physical effects (e.g., occlusion, warmth, mechanical cleansing, or placebo). A future study employing a matched vehicle patch without the okra extract is necessary to isolate the pharmacological effect. Third, the concurrent use of meibomian gland massage in both groups, while reflecting standard adjunctive care, precludes isolating the specific effect of either the okra eye patch or the control eye drops. As massage itself improves MGD, a portion of the observed benefit in both groups is likely attributable to this common background therapy. Consequently, the observed improvements should be interpreted as the result of the combined regimen. Future studies using a factorial design (e.g., including a massage-only group) are needed to quantify the incremental benefit of the primary interventions. Fourth, the administration methods of okra eye patches and eye drops are different. Therefore, it remains unclear whether the therapeutic benefits are attributable primarily to the bioactive components of okra or to the physical cleansing effect of the patch application. Most importantly, this clinical study cannot define the precise active components or the dominant mechanism of action (e.g., anti-inflammatory, antioxidant, or mechanical) of the okra patch. To address these uncertainties, future preclinical studies utilizing animal models of dry eye are warranted, employing standardized okra extracts in a controlled formulation (e.g., eye drops) to investigate dose-response relationships and underlying molecular pathways. Fifth, the single-center design and limited sample size necessitate validation in larger, multi-center trials. No a priori sample size calculation was performed; therefore, this study should be considered exploratory/pilot in nature. Unlike IPL or Lipiflow, the okra eye patch offers a low-cost, home-based option suitable for mild-to-moderate MGD or maintenance therapy. Conclusion In conclusion, this single-center pilot study provides preliminary evidence that the okra eye patch may be a promising and well-tolerated intervention for MGD associated with dry eye. It demonstrated improvements in patient symptoms and signs, alongside objective enhancements in lipid layer thickness, tear film stability, corneal epithelial integrity, and the quality of meibomian gland acini. The overall treatment effect, assessed through a multidimensional evaluation including symptoms, signs, ocular surface interferometry, OQAS II system and IVCM, showed trends comparable to sodium hyaluronate combined with ofloxacin eye drops. Given the small sample size and open-label design, these preliminary findings should be interpreted with caution and require confirmation through larger, double-masked, sham-controlled trials.