PURPOSE: To classify subtypes of meibomian gland dysfunction (MGD) and evaluate the dependency of dry eye signs, symptoms, and parameters on MGD subtype. DESIGN: Cross-sectional study. STUDY POPULATION: the right eyes of 447 patients with MGD of various subtypes and 20 healthy volunteers. METHODS: Patients were divided into 4 subtypes of MGD based on meibum expression, meibum quality, and MG loss on meibography images (meibograde of 0-6). Subtypes were patients with high meibum delivery (hypersecretory and nonobvious MGD) and those with low meibum delivery (hyposecretory and obstructive MGD). Additional clinical tests included tear film break-up time (TFBUT), ocular staining, osmolarity, Schirmer I, blink interval timing and the Ocular Surface Disease Index (OSDI) questionnaire. RESULTS: A total of 78 eyes had hypersecretory MGD; 49 eyes had nonobvious MGD; 66 eyes had hyposecretory MGD; and 254 eyes had obstructive MGD. Increased tear film osmolarity and lower TFBUT were found in the low-delivery groups; hyposecretory (P = 0.006, P = 0.016) and obstructive MGD (P = 0.008, P = 0.006) relative to high-delivery MGD (hypersecretory and nonobvious groups, respectively). Worse ocular symptoms and ocular staining were also found in low-delivery MGD groups than the high delivery MGD groups (P < 0.01 and P < 0.006, respectively). CONCLUSIONS: Patients with low-delivery MGD had worse dry eye parameters and ocular symptoms than those with high meibum delivery, indicating the pivotal role of meibum secretion in ocular surface health that should be targeted in MGD therapy. Furthermore, nonobvious MGD cannot be diagnosed using conventional dry eye tests and requires morphologic assessment of meibography images to confirm MG loss.

Deep learning (DL) is a subset of artificial intelligence (AI), which uses multilayer neural networks modelled after the mammalian visual cortex capable of synthesizing images in ways that will transform the field of glaucoma. Autonomous DL algorithms are capable of maximizing information embedded in digital fundus photographs and ocular coherence tomographs to outperform ophthalmologists in disease detection. Other unsupervised algorithms such as principal component analysis (axis learning) and archetypal analysis (corner learning) facilitate visual field interpretation and show great promise to detect functional glaucoma progression and differentiate it from non-glaucomatous changes when compared with conventional software packages. Forecasting tools such as the Kalman filter may revolutionize glaucoma management by accounting for a host of factors to set target intraocular pressure goals that preserve vision. Activation maps generated from DL algorithms that process glaucoma data have the potential to efficiently direct our attention to critical data elements embedded in high throughput data and enhance our understanding of the glaucomatous process. It is hoped that AI will realize more accurate assessment of the copious data encountered in glaucoma management, improving our understanding of the disease, preserving vision, and serving to enhance the deep bonds that patients develop with their treating physicians.

PURPOSE: To investigate whether anti-vascular endothelial growth factor (anti-VEGF) for diabetic macular edema or proliferative diabetic retinopathy (PDR) increases the risk of traction retinal detachment (TRD) among eyes with PDR. METHODS: Pooled analysis of PDR eyes from Protocols I, J, N, S, or T with Early Treatment Diabetic Retinopathy Study level ≥61 (prompt vitrectomy was not planned) randomly assigned to the control group (laser photocoagulation, sham, or intravitreal saline; 396 eyes) or anti-VEGF (487 eyes). The primary outcome was investigator-identified TRD within 1 year of randomization. RESULTS: The 1-year cumulative probability of TRD was 6.8% (95% confidence interval: 4.6%-9.9%, 25 events) in control-group eyes and 4.8% (95% confidence interval: 3.2%-7.3%, 22 events) in anti-VEGF group eyes (hazard ratio = 0.95 [95% confidence interval: 0.54-1.66, P = 0.86]). The cumulative probability of vitrectomy for TRD was 4.4% (16 events) in control-group eyes and 2.2% (9 events) in anti-VEGF group eyes (P = 0.19). Percentage with TRD and vitrectomy for TRD were similar within strata of diabetic retinopathy severity. CONCLUSION: These findings do not support the hypothesis that anti-VEGF therapy for diabetic macular edema or PDR increases the risk of TRD among eyes with PDR similar to those enrolled in five DRCR Retina Network protocols for which prompt vitrectomy was not planned.

PURPOSE: To describe the etiology and clinical characteristics of macular edema (ME) in patients with familial exudative vitreoretinopathy. METHODS: Observational, retrospective case series of 30 patients (34 eyes) with ME and familial exudative vitreoretinopathy who underwent spectral-domain optical coherence tomography imaging between 2009 and 2016. Baseline and follow-up optical coherence tomographies were correlated with color fundus photography and fluorescein angiography. RESULTS: The average age was 20.6 years (6.6-68.7). Eighteen eyes exhibited cystoid ME (52.9%), 14 noncystoid ME (41.2%), and 2 eyes (5.9%) with both. Macular edema was foveal in 52.9% (n = 18). Eighteen of 24 eyes (64.3%) with an available fluorescein angiography showed leakage from ME. The most common structural feature was posterior hyaloidal organization/contraction (n = 15). Sixteen eyes were treated with topical or intravitreal steroids (n = 6), intravitreal anti-vascular endothelial growth factor (n = 3), or pars plana vitrectomy with membrane stripping (n = 7). There was no difference between mean preoperative and postoperative LogMAR visual acuity (0.63 [20/85] vs. 0.87 [20/148], P = 0.35) after vitrectomy despite a statistical improvement in the mean central foveal thickness (596 mm vs. 303 mm, P = 0.04). CONCLUSION: Macular edema in familial exudative vitreoretinopathy occurs most commonly because of traction. Vitrectomy is effective for relieving tractional forces with anatomical improvement.

PURPOSE: To use swept-source optical coherence tomography and swept-source optical coherence tomography angiography to investigate potential relationships between choroidal vascular hyperpermeability (CVH) seen with indocyanine green angiography (ICGA), choriocapillaris flow density, and choroidal thickness in eyes with pachychoroid pigment epitheliopathy. METHODS: Patients with pachychoroid pigment epitheliopathy were prospectively imaged with 12-mm × 12-mm swept-source optical coherence tomography, 12-mm × 12-mm swept-source optical coherence tomography angiographyA, and ICGA. Binarized choriocapillaris OCTA images were superimposed with ICGA images in which CVH area had been isolated. Choriocapillaris flow density within or outside the quadrants of CVH was calculated and the ratio of these two values was determined. The presence of CVH and choroidal thickness was evaluated at 9 locations within a central 3-mm × 3-mm area to explore the relationship between these 2 factors. RESULTS: Ten eyes from 10 patients were enrolled in the present study. Choriocapillaris flow density within quadrants of CVH area was significantly lower compared with quadrants without CVH (P < 0.001). The mean choriocapillaris flow density ratio was 0.86 ± 0.10 (range: 0.65-0.99). From among the 90 locations in 10 study eyes, 48 were within areas of CVH. Choroidal thickness was greater in quadrants of CVH compared with areas without CVH (P < 0.001, 455 ± 122 µm vs. 297 ± 93 µm). CONCLUSION: Reduced choriocapillaris flow density, increased choroidal thickness, and CVH appear to co-localize in eyes with pachychoroid pigment epitheliopathy.

PURPOSE: To determine whether small macular hole closure can be achieved with 25-G vitrectomy surgery with internal limiting membrane peeling without the use of intraocular gas tamponade or facedown positioning. METHODS: 25-G vitrectomy surgery with internal limiting membrane peeling without the use of intraocular gas tamponade or positioning was performed on 20 eyes with a small (<400-µm diameter), full-thickness macular hole. RESULTS: In 17 of 20 eyes (85%), the hole had closed. Three holes had closed by Postoperative Day 1, 13 holes by Postoperative Week 1, 16 holes by Postoperative Week 2, and 17 holes by Postoperative Week 6. At Postoperative Month 1, vision improved in 16 of 17 eyes in which the macular hole had closed. One hole that had not closed at the first postoperative week and two holes that had not closed at the third postoperative week required follow-up surgery with intraocular gas tamponade and facedown positioning, after which the hole closed. The mean preoperative visual acuity was 0.626 logMAR (20/85), and the mean postoperative visual acuity after 1 month was 0.392 logMAR (20/50) (P < 0.001). CONCLUSION: Vitrectomy surgery with internal limiting membrane peeling without the use of gas tamponade or positioning can achieve closure of small macular holes.

: Uveitis is an important cause of blindness and ocular morbidity in the world. The patterns of uveitis have not been well characterized in sub-Saharan Africa. : To describe the characteristics of uveitis among patients presenting to Jimma University Department of Ophthalmology (JUDO) from July 2013 to December 2014. : This hospital-based prospective cross-sectional study included all new uveitis patients visiting JUDO outpatient department during the study period. : Among 98 patients diagnosed with uveitis, anterior uveitis was found in 74.5% of patients. Majority of the patients, 83.7%, had unilateral uveitis. A uveitis syndrome was identified in 22.5% of cases; of these 15 (68.2%) were infectious. Herpes simplex uveitis was the commonest infectious cause (53.3%) while Toxoplasmosis was the most common cause of posterior uveitis (60%). : Anterior uveitis was the most common pattern found among uveitis patients. Herpes simplex and toxoplasmic chorioretinitis were the most common-identified infectious causes.

The glycocalyx is a dense and diverse coat of glycans and glycoconjugates responsible for maintaining cell surface integrity and regulating the interaction of cells with the external environment. Transmembrane mucins such as MUC1 and MUC16 comprise a major component of the epithelial glycocalyx and are currently used to monitor disease progression in cancer. At the ocular surface, multiple lines of evidence indicate that abnormal expression of the enzymes responsible for glycan biosynthesis during pathological conditions impairs the glycosylation of transmembrane mucins. It is now becoming clear that these changes contribute to modify the interaction of mucins with galectin-3, a multimeric lectin crucial for preserving the ocular surface epithelial barrier. This review highlights the potential of using the epithelial glycocalyx as a reliable source for the generation of biomarkers to diagnose and monitor ocular surface disease.

BACKGROUND: The genetic basis of monocular elevation deficiency (MED) is unclear. It has previously been considered to arise due to a supranuclear abnormality. METHODS: Two brothers with MED were referred to Leicester Royal Infirmary, UK from the local opticians. Their father had bilateral ptosis and was unable to elevate both eyes, consistent with the diagnosis of congenital fibrosis of extraocular muscles (CFEOM). Candidate sequencing was performed in all family members. RESULTS: Both affected siblings (aged 7 and 12 years) were unable to elevate the right eye. Their father had bilateral ptosis, left esotropia and bilateral limitation of elevation. Chin up head posture was present in the older sibling and the father. Bell's phenomenon and vertical rotational vestibulo-ocular reflex were absent in the right eye for both children. Mild bilateral facial nerve palsy was present in the older sibling and the father. Both siblings had slight difficulty with tandem gait. MRI revealed hypoplastic oculomotor nerve. Left anterior insular focal cortical dysplasia was seen in the older sibling. Sequencing of revealed a novel heterozygous variant (c.1263G>C, p.E421D) segregating with the phenotype. This residue is in the C-terminal H12 α-helix of β-tubulin and is one of three putative kinesin binding sites. CONCLUSION: We show that familial MED can arise from a variant and could be considered a limited form of CFEOM. Neurological features such as mild facial palsy and cortical malformations can be present in patients with MED. Thus, in individuals with congenital MED, consideration may be made for mutation screening.