Morphological characteristics of Meibomian Glands and their Influence on Dry Eye disease in contact lens wearers

LipiFlow for the treatment of dry eye disease: A Cochrane systematic review summary

PURPOSE

To evaluate the effectiveness and the safety of LipiFlow for treating signs and symptoms of dry eye disease (DED) in adults.

METHODS

The following databases were searched for randomized trials: CENTRAL, MEDLINE Ovid, Embase.com, PubMed, LILACS, ClinicalTrials.gov, and WHO ICTRP on 24 October 2022. The included studies were conducted in adults (≥18 years) with DED or meibomian gland dysfunction (MGD) as defined by the investigators. Standard Cochrane methodology was applied.

RESULTS

This study included 13 trials that randomized a total of 1,155 participants (66 % female; age range = 19 to 86 years). Five trials compared LipiFlow with basic warm compresses. Analyzing symptom scores in these trials yielded conflicting evidence of a difference in symptoms between LipiFlow and basic warm compresses after 4 weeks. There was no evidence of a difference in meibomian gland expression, meibum quality, or tear breakup time when comparing LipiFlow with basic warm compresses. Another 5 trials compared LipiFlow with thermostatic devices. Analysis of symptom scores in these trials at 4 weeks showed that thermostatic devices had reduced Ocular Surface Disease Index (OSDI) scores by a mean difference of 4.59 as compared with LipiFlow. The remaining 3 included trials could not be grouped for comparisons. The overall evidence was of low or very low certainty, with most trials being assessed as having a high risk of bias. No trial reported any intervention-related, vision-threating adverse events.

CONCLUSIONS

LipiFlow performs similarly to other DED treatments. Further research with adequate masking and a standardized testing methodology is still needed.

Comparing meibomian gland visibility on optical coherence tomography and Keratograph 5M images using objective and subjective grading methods

PURPOSE

To investigate if there is a visible difference in meibomian gland (MG) length between images captured with the Visante optical coherence tomography (OCT; wavelength = 1,310 nm) and the OCULUS Keratograph 5M (K5M; wavelength = 880 nm).

METHODS

Adults between 18 and 40 years were recruited. Baseline dry eye disease was evaluated with the Standard Patient Evaluation of Eye Dryness (SPEED) and tear meniscus height and tear breakup time with the K5M. Right upper and lower eyelid MGs were imaged with the K5M and Visante OCT. Each image was graded with the 0 to 3 meiboscore scale. The central 5 MGs were evaluated with ImageJ for percent gland length visibility.

RESULTS

Thirty participants were analyzed with a median (interquartile range [IQR]) age of 23.0 (5.0) years (53.3 % female). Overall, participants were asymptomatic and had normal tear films. Meiboscores based on K5M and Visante OCT was significantly different for the lower eyelid (0[1] vs 1[2]; p = 0.007) but not the upper eyelid (0[1] vs 0[1]; p = 1.00). The mean percent gland visibility of the upper eyelid (82.7[9.6] vs 75.2[13.5]; p < 0.001) and the lower eyelid (81.2[12.7] vs 64.1[17.6]; p < 0.001) were significantly greater on the Visante OCT than the K5M images, respectively.

CONCLUSION

OCT images had significantly greater percent visible MG lengths than the K5M images. This suggests viable segments of the MGs may be missed with typical imaging, which may explain how it is possible that studies have found less post-treatment MG atrophy.

Unsupervised Learning Based on Meibography Enables Subtyping of Dry Eye Disease and Reveals Ocular Surface Features

Purpose

This study aimed to establish an image-based classification that can reveal the clinical characteristics of patients with dry eye using unsupervised learning methods.

Methods

In this study, we analyzed 82,236 meibography images from 20,559 subjects. Using the SimCLR neural network, the images were categorized. Data for each patient were averaged and subjected to mini-batch k-means clustering, and validated through consensus clustering. Statistical metrics determined optimal category numbers. Using a UNet model, images were segmented to identify meibomian gland (MG) areas. Clinical features were assessed, including tear breakup time (BUT), tear meniscus height (TMH), and gland atrophy. A thorough ocular surface evaluation was conducted on 280 cooperative patients.

Results

SimCLR neural network achieved clustering patients with dry eye into six image-based subtypes. Patients in different subtypes harbored significantly different noninvasive BUT, significantly correlated with TMH. Subtypes 1 and 5 had the most severe MG atrophy. Subtype 2 had the highest corneal fluorescent staining (CFS). Subtype 4 had the lowest TMH, whereas subtype 5 had the highest. Subtypes 3 and 6 had the largest MG areas, and the upper MG areas of a person's bilateral eyes were highly correlated. Image-based subtypes are related to meibum quality, CFS, and morphological characteristics of MG.

Conclusions

In this study, we developed an unsupervised neural network model to cluster patients with dry eye into image-based subtypes using meibography images. We annotated these subtypes with functional and morphological clinical characteristics.

Association of meibomian gland morphology with orifice plugging and lid margin thickening in meibomian gland dysfunction patients

PURPOSE

We sought to investigate the association of meibomian gland morphology with lid margin abnormalities in patients with meibomian gland dysfunction.

METHODS

This retrospective study included 368 eyes of 184 patients. Meibography was used to evaluate meibomian gland (MG) morphological features, such as dropout, distortion, thickened ratio and thinned ratio. Lid margin photography was used to evaluate lid margin abnormalities including orifice plugging, vascularity, irregularity and thickening. The association between MG morphological features and lid margin abnormalities was analyzed using a mixed linear model.

RESULTS

The study found a positive correlation between plugging of gland orifices grade and MG dropout grade in both the upper lids (B = 0.40, p = 0.007) and the lower lids (B = 0.55, p = 0.001). Plugging of gland orifices grade was also positively correlated to MG distortion grade in the upper lids (B = 0.75, p = 0.006). In the upper lids, MG thickened ratio increased first (B = 0.21, p = 0.003) and then decreased (B = - 0.14, p = 0.010) with a higher lid margin thickening grade. MG thinned ratio was negatively correlated with lid margin thickening (B = - 0.14, p = 0.002, B = - 0.13, p = 0.007). MG distortion grade decreased with lid margin thickening (B = - 0.61, p = 0.012).

CONCLUSION

Orifice plugging was correlated to meibomian gland distortion and dropout. Lid margin thickening was associated with meibomian gland thickened ratio, thinned ratio, and distortion. The study also suggested that distorted and thinned glands may be transitional phases between thickened glands and glands dropout.

TFOS Lifestyle: Impact of lifestyle challenges on the ocular surface

Many factors in the domains of mental, physical, and social health have been associated with various ocular surface diseases, with most of the focus centered on aspects of dry eye disease (DED). Regarding mental health factors, several cross-sectional studies have noted associations between depression and anxiety, and medications used to treat these disorders, and DED symptoms. Sleep disorders (both involving quality and quantity of sleep) have also been associated with DED symptoms. Under the domain of physical health, several factors have been linked to meibomian gland abnormalities, including obesity and face mask wear. Cross-sectional studies have also linked chronic pain conditions, specifically migraine, chronic pain syndrome and fibromyalgia, to DED, principally focusing on DED symptoms. A systematic review and meta-analysis reviewed available data and concluded that various chronic pain conditions increased the risk of DED (variably defined), with odds ratios ranging from 1.60 to 2.16. However, heterogeneity was noted, highlighting the need for additional studies examining the impact of chronic pain on DED signs and subtype (evaporative versus aqueous deficient). With respect to societal factors, tobacco use has been most closely linked to tear instability, cocaine to decreased corneal sensitivity, and alcohol to tear film disturbances and DED symptoms.

Automated quantification of meibomian gland dropout in infrared meibography using deep learning

PURPOSE

Develop a deep learning-based automated method to segment meibomian glands (MG) and eyelids, quantitatively analyze the MG area and MG ratio, estimate the meiboscore, and remove specular reflections from infrared images.

METHODS

A total of 1600 meibography images were captured in a clinical setting. 1000 images were precisely annotated with multiple revisions by investigators and graded 6 times by meibomian gland dysfunction (MGD) experts. Two deep learning (DL) models were trained separately to segment areas of the MG and eyelid. Those segmentation were used to estimate MG ratio and meiboscores using a classification-based DL model. A generative adversarial network was implemented to remove specular reflections from original images.

RESULTS

The mean ratio of MG calculated by investigator annotation and DL segmentation was consistent 26.23% vs 25.12% in the upper eyelids and 32.34% vs. 32.29% in the lower eyelids, respectively. Our DL model achieved 73.01% accuracy for meiboscore classification on validation set and 59.17% accuracy when tested on images from independent center, compared to 53.44% validation accuracy by MGD experts. The DL-based approach successfully removes reflection from the original MG images without affecting meiboscore grading.

CONCLUSIONS

DL with infrared meibography provides a fully automated, fast quantitative evaluation of MG morphology (MG Segmentation, MG area, MG ratio, and meiboscore) which are sufficiently accurate for diagnosing dry eye disease. Also, the DL removes specular reflection from images to be used by ophthalmologists for distraction-free assessment.