FAVOR-GPT: a generative natural language interface to whole genome variant functional annotations.

Motivation: Functional Annotation of genomic Variants Online Resources (FAVOR) offers multi-faceted, whole genome variant functional annotations, which is essential for Whole Genome and Exome Sequencing (WGS/WES) analysis and the functional prioritization of disease-associated variants. A versatile chatbot designed to facilitate informative interpretation and interactive, user-centric summary of the whole genome variant functional annotation data in the FAVOR database is needed. Results: We have developed FAVOR-GPT, a generative natural language interface powered by integrating large language models (LLMs) and FAVOR. It is developed based on the Retrieval Augmented Generation (RAG) approach, and complements the original FAVOR portal, enhancing usability for users, especially those without specialized expertise. FAVOR-GPT simplifies raw annotations by providing interpretable explanations and result summaries in response to the user's prompt. It shows high accuracy when cross-referencing with the FAVOR database, underscoring the robustness of the retrieval framework. Availability and implementation: Researchers can access FAVOR-GPT at FAVOR's main website (https://favor.genohub.org).

Single-cell RNA sequencing reveals placental response under environmental stress.

The placenta is crucial for fetal development, yet the impact of environmental stressors such as arsenic exposure remains poorly understood. We apply single-cell RNA sequencing to analyze the response of the mouse placenta to arsenic, revealing cell-type-specific gene expression, function, and pathological changes. Notably, the Prap1 gene, which encodes proline-rich acidic protein 1 (PRAP1), is significantly upregulated in 26 placental cell types including various trophoblast cells. Our study shows a female-biased increase in PRAP1 in response to arsenic and localizes it in the placenta. In vitro and ex vivo experiments confirm PRAP1 upregulation following arsenic treatment and demonstrate that recombinant PRAP1 protein reduces arsenic-induced cytotoxicity and downregulates cell cycle pathways in human trophoblast cells. Moreover, PRAP1 knockdown differentially affects cell cycle processes, proliferation, and cell death depending on the presence of arsenic. Our findings provide insights into the placental response to environmental stress, offering potential preventative and therapeutic approaches for environment-related adverse outcomes in mothers and children.

Whole genome sequencing based analysis of inflammation biomarkers in the Trans-Omics for Precision Medicine (TOPMed) consortium.

Inflammation biomarkers can provide valuable insight into the role of inflammatory processes in many diseases and conditions. Sequencing based analyses of such biomarkers can also serve as an exemplar of the genetic architecture of quantitative traits. To evaluate the biological insight, which can be provided by a multi-ancestry, whole-genome based association study, we performed a comprehensive analysis of 21 inflammation biomarkers from up to 38 465 individuals with whole-genome sequencing from the Trans-Omics for Precision Medicine (TOPMed) program (with varying sample size by trait, where the minimum sample size was n = 737 for MMP-1). We identified 22 distinct single-variant associations across 6 traits-E-selectin, intercellular adhesion molecule 1, interleukin-6, lipoprotein-associated phospholipase A2 activity and mass, and P-selectin-that remained significant after conditioning on previously identified associations for these inflammatory biomarkers. We further expanded upon known biomarker associations by pairing the single-variant analysis with a rare variant set-based analysis that further identified 19 significant rare variant set-based associations with 5 traits. These signals were distinct from both significant single variant association signals within TOPMed and genetic signals observed in prior studies, demonstrating the complementary value of performing both single and rare variant analyses when analyzing quantitative traits. We also confirm several previously reported signals from semi-quantitative proteomics platforms. Many of these signals demonstrate the extensive allelic heterogeneity and ancestry-differentiated variant-trait associations common for inflammation biomarkers, a characteristic we hypothesize will be increasingly observed with well-powered, large-scale analyses of complex traits.

A statistical framework for powerful multi-trait rare variant analysis in large-scale whole-genome sequencing studies.

Large-scale whole-genome sequencing (WGS) studies have improved our understanding of the contributions of coding and noncoding rare variants to complex human traits. Leveraging association effect sizes across multiple traits in WGS rare variant association analysis can improve statistical power over single-trait analysis, and also detect pleiotropic genes and regions. Existing multi-trait methods have limited ability to perform rare variant analysis of large-scale WGS data. We propose MultiSTAAR, a statistical framework and computationally-scalable analytical pipeline for functionally-informed multi-trait rare variant analysis in large-scale WGS studies. MultiSTAAR accounts for relatedness, population structure and correlation among phenotypes by jointly analyzing multiple traits, and further empowers rare variant association analysis by incorporating multiple functional annotations. We applied MultiSTAAR to jointly analyze three lipid traits (low-density lipoprotein cholesterol, high-density lipoprotein cholesterol and triglycerides) in 61,861 multi-ethnic samples from the Trans-Omics for Precision Medicine (TOPMed) Program. We discovered new associations with lipid traits missed by single-trait analysis, including rare variants within an enhancer of NIPSNAP3A and an intergenic region on chromosome 1.

Functional characterization of Alzheimer's disease genetic variants in microglia.

Candidate cis-regulatory elements (cCREs) in microglia demonstrate the most substantial enrichment for Alzheimer's disease (AD) heritability compared to other brain cell types. However, whether and how these genome-wide association studies (GWAS) variants contribute to AD remain elusive. Here we prioritize 308 previously unreported AD risk variants at 181 cCREs by integrating genetic information with microglia-specific 3D epigenome annotation. We further establish the link between functional variants and target genes by single-cell CRISPRi screening in microglia. In addition, we show that AD variants exhibit allelic imbalance on target gene expression. In particular, rs7922621 is the effective variant in controlling TSPAN14 expression among other nominated variants in the same cCRE and exerts multiple physiological effects including reduced cell surface ADAM10 and altered soluble TREM2 (sTREM2) shedding. Our work represents a systematic approach to prioritize and characterize AD-associated variants and provides a roadmap for advancing genetic association to experimentally validated cell-type-specific phenotypes and mechanisms.

Whole Genome Sequencing Based Analysis of Inflammation Biomarkers in the Trans-Omics for Precision Medicine (TOPMed) Consortium.

Inflammation biomarkers can provide valuable insight into the role of inflammatory processes in many diseases and conditions. Sequencing based analyses of such biomarkers can also serve as an exemplar of the genetic architecture of quantitative traits. To evaluate the biological insight, which can be provided by a multi-ancestry, whole-genome based association study, we performed a comprehensive analysis of 21 inflammation biomarkers from up to 38,465 individuals with whole-genome sequencing from the Trans-Omics for Precision Medicine (TOPMed) program. We identified 22 distinct single-variant associations across 6 traits - E-selectin, intercellular adhesion molecule 1, interleukin-6, lipoprotein-associated phospholipase A2 activity and mass, and P-selectin - that remained significant after conditioning on previously identified associations for these inflammatory biomarkers. We further expanded upon known biomarker associations by pairing the single-variant analysis with a rare variant set-based analysis that further identified 19 significant rare variant set-based associations with 5 traits. These signals were distinct from both significant single variant association signals within TOPMed and genetic signals observed in prior studies, demonstrating the complementary value of performing both single and rare variant analyses when analyzing quantitative traits. We also confirm several previously reported signals from semi-quantitative proteomics platforms. Many of these signals demonstrate the extensive allelic heterogeneity and ancestry-differentiated variant-trait associations common for inflammation biomarkers, a characteristic we hypothesize will be increasingly observed with well-powered, large-scale analyses of complex traits.

Genetic regulators of sputum mucin concentration and their associations with COPD phenotypes.

Hyper-secretion and/or hyper-concentration of mucus is a defining feature of multiple obstructive lung diseases, including chronic obstructive pulmonary disease (COPD). Mucus itself is composed of a mixture of water, ions, salt and proteins, of which the gel-forming mucins, MUC5AC and MUC5B, are the most abundant. Recent studies have linked the concentrations of these proteins in sputum to COPD phenotypes, including chronic bronchitis (CB) and acute exacerbations (AE). We sought to determine whether common genetic variants influence sputum mucin concentrations and whether these variants are also associated with COPD phenotypes, specifically CB and AE. We performed a GWAS to identify quantitative trait loci for sputum mucin protein concentration (pQTL) in the Sub-Populations and InteRmediate Outcome Measures in COPD Study (SPIROMICS, n = 708 for total mucin, n = 215 for MUC5AC, MUC5B). Subsequently, we tested for associations of mucin pQTL with CB and AE using regression modeling (n = 822-1300). Replication analysis was conducted using data from COPDGene (n = 5740) and by examining results from the UK Biobank. We identified one genome-wide significant pQTL for MUC5AC (rs75401036) and two for MUC5B (rs140324259, rs10001928). The strongest association for MUC5B, with rs140324259 on chromosome 11, explained 14% of variation in sputum MUC5B. Despite being associated with lower MUC5B, the C allele of rs140324259 conferred increased risk of CB (odds ratio (OR) = 1.42; 95% confidence interval (CI): 1.10-1.80) as well as AE ascertained over three years of follow up (OR = 1.41; 95% CI: 1.02-1.94). Associations between rs140324259 and CB or AE did not replicate in COPDGene. However, in the UK Biobank, rs140324259 was associated with phenotypes that define CB, namely chronic mucus production and cough, again with the C allele conferring increased risk. We conclude that sputum MUC5AC and MUC5B concentrations are associated with common genetic variants, and the top locus for MUC5B may influence COPD phenotypes, in particular CB.

ATR inhibition overcomes platinum tolerance associated with ERCC1- and p53-deficiency by inducing replication catastrophe.

ERCC1/XPF is a heterodimeric DNA endonuclease critical for repair of certain chemotherapeutic agents. We recently identified that ERCC1- and p53-deficient lung cancer cells are tolerant to platinum-based chemotherapy. ATR inhibition synergistically re-stored platinum sensitivity to platinum tolerant ERCC1-deficient cells. Mechanistically we show this effect is reliant upon several functions of ATR including replication fork protection and altered cell cycle checkpoints. Utilizing an inhibitor of replication protein A (RPA), we further demonstrate that replication fork protection and RPA availability are critical for platinum-based drug tolerance. Dual treatment led to increased formation of DNA double strand breaks and was associated with chromosome pulverization. Combination treatment was also associated with increased micronuclei formation which were capable of being bound by the innate immunomodulatory factor, cGAS, suggesting that combination platinum and ATR inhibition may also enhance response to immunotherapy in ERCC1-deficient tumors. In vivo studies demonstrate a significant effect on tumor growth delay with combination therapy compared with single agent treatment. Results of this study have led to the identification of a feasible therapeutic strategy combining ATR inhibition with platinum and potentially immune checkpoint blockade inhibitors to overcome platinum tolerance in ERCC1-deficient, p53-mutant lung cancers.

INCREASED DISTANCE FROM THE CLINIC LEADS TO HIGHER LOSS TO FOLLOW-UP AFTER PARS PLANA VITRECTOMY IN DIABETIC PATIENTS.

PURPOSE: The importance of consistent outpatient follow-up for management of diabetic eye disease has been well-established. The objective of this study was to identify patient factors associated with being lost to follow-up in postsurgical patients after undergoing pars plana vitrectomy for diabetic eye disease. METHODS: The charts of diabetic patients undergoing pars plana vitrectomy for nonclearing vitreous hemorrhage at an academic medical center by a single surgeon between 2012 and 2019 were reviewed. The rates of loss to follow-up during the postoperative period were compared based on patient distance from the clinic and insurance status. RESULTS: A total of 144 patients met inclusion criteria. A total of 45 patients (31.25%) were lost to follow-up during the 3-month postoperative period. The rate of loss to follow-up increased with every postoperative visit and was significantly higher for patients living greater than 30 miles from the clinic versus patients living within 30 miles from the clinic. There was no statistically significant difference in loss to follow-up based on insurance status. CONCLUSION: Increased distance from the clinic presents a challenge to providing safe and effective postsurgical care to diabetic patients. This presents opportunities for comanagement or other creative strategies to improve postsurgical follow-up rates for at-risk patients.

IL-33 enhances Jagged1 mediated NOTCH1 intracellular domain (NICD) deubiquitination and pathological angiogenesis in proliferative retinopathy.

Pathological retinal neovascularization (NV) is a clinical manifestation of various proliferative retinopathies, and treatment of NV using anti-VEGF therapies is not selective, as it also impairs normal retinal vascular growth and function. Here, we show that genetic deletion or siRNA-mediated downregulation of IL-33 reduces pathological NV in a murine model of oxygen-induced retinopathy (OIR) with no effect on the normal retinal repair. Furthermore, our fluorescent activated cell sorting (FACS) data reveals that the increase in IL-33 expression is in endothelial cells (ECs) of the hypoxic retina and conditional genetic deletion of IL-33 in retinal ECs reduces pathological NV. In vitro studies using human retinal microvascular endothelial cells (HRMVECs) show that IL-33 induces sprouting angiogenesis and requires NFkappaB-mediated Jagged1 expression and Notch1 activation. Our data also suggest that IL-33 enhances de-ubiquitination and stabilization of Notch1 intracellular domain via its interaction with BRCA1-associated protein 1 (BAP1) and Numb in HRMVECs and a murine model of OIR.

TWO-SIGMA-G: a new competitive gene set testing framework for scRNA-seq data accounting for inter-gene and cell-cell correlation.

We propose TWO-SIGMA-G, a competitive gene set test for scRNA-seq data. TWO-SIGMA-G uses a mixed-effects regression model based on our previously published TWO-SIGMA to test for differential expression at the gene-level. This regression-based model provides flexibility and rigor at the gene-level in (1) handling complex experimental designs, (2) accounting for the correlation between biological replicates and (3) accommodating the distribution of scRNA-seq data to improve statistical inference. Moreover, TWO-SIGMA-G uses a novel approach to adjust for inter-gene-correlation (IGC) at the set-level to control the set-level false positive rate. Simulations demonstrate that TWO-SIGMA-G preserves type-I error and increases power in the presence of IGC compared with other methods. Application to two datasets identified HIV-associated interferon pathways in xenograft mice and pathways associated with Alzheimer's disease progression in humans.

TWO-SIGMA: A novel two-component single cell model-based association method for single-cell RNA-seq data.

In this paper, we develop TWO-SIGMA, a TWO-component SInGle cell Model-based Association method for differential expression (DE) analyses in single-cell RNA-seq (scRNA-seq) data. The first component models the probability of "drop-out" with a mixed-effects logistic regression model and the second component models the (conditional) mean expression with a mixed-effects negative binomial regression model. TWO-SIGMA is extremely flexible in that it: (i) does not require a log-transformation of the outcome, (ii) allows for overdispersed and zero-inflated counts, (iii) accommodates a correlation structure between cells from the same individual via random effect terms, (iv) can analyze unbalanced designs (in which the number of cells does not need to be identical for all samples), (v) can control for additional sample-level and cell-level covariates including batch effects, (vi) provides interpretable effect size estimates, and (vii) enables general tests of DE beyond two-group comparisons. To our knowledge, TWO-SIGMA is the only method for analyzing scRNA-seq data that can simultaneously accomplish each of these features. Simulations studies show that TWO-SIGMA outperforms alternative regression-based approaches in both type-I error control and power enhancement when the data contains even moderate within-sample correlation. A real data analysis using pancreas islet single-cells exhibits the flexibility of TWO-SIGMA and demonstrates that incorrectly failing to include random effect terms can have dramatic impacts on scientific conclusions. TWO-SIGMA is implemented in the R package twosigma available at https://github.com/edvanburen/twosigma.

The miR-183/96/182 Cluster Regulates the Functions of Corneal Resident Macrophages.

Tissue-resident macrophages (ResMϕ) play important roles in the normal development and physiological functions as well as tissue repair and immune/inflammatory response to both internal and external insults. In cornea, ResMϕ are critical to the homeostasis and maintenance, wound healing, ocular immune privilege, and immune/inflammatory response to injury and microbial infection. However, the roles of microRNAs in corneal ResMϕ are utterly unknown. Previously, we demonstrated that the conserved miR-183/96/182 cluster (miR-183/96/182) plays important roles in sensory neurons and subgroups of both innate and adaptive immune cells and modulates corneal response to bacterial infection. In this study, we provide direct evidence that the mouse corneal ResMϕ constitutively produce both IL-17f and IL-10. This function is regulated by miR-183/96/182 through targeting Runx1 and Maf, key transcriptional regulators for IL-17f and IL-10 expression, respectively. In addition, we show that miR-183/96/182 has a negative feedback regulation on the TLR4 pathway in mouse corneal ResMϕ. Furthermore, miR-183/96/182 regulates the number of corneal ResMϕ. Inactivation of miR-183/96/182 in mouse results in more steady-state corneal resident immune cells, including ResMϕ, and leads to a simultaneous early upregulation of innate IL-17f and IL-10 production in the cornea after Pseudomonas aeruginosa infection. Its multiplex regulations on the simultaneous production of IL-17f and IL-10, TLR4 signaling pathway and the number of corneal ResMϕ place miR-183/96/182 in the center of corneal innate immunity, which is key to the homeostasis of the cornea, ocular immune privilege, and the corneal response to microbial infections.

The Relationship of Travel Distance to Postoperative Follow-up Care on Glaucoma Surgery Outcomes.

PRéCIS:: This study addresses the paucity of literature examining glaucoma patients' distance from clinic on postoperative follow-up outcomes. Greater distance from clinic was associated with higher likelihood of loss to follow-up and missed appointments. PURPOSE: To investigate the relationship of patient travel distance and interstate access to glaucoma surgery postoperative follow-up visit attendance. METHODS AND PARTICIPANTS: Retrospective longitudinal chart review of all noninstitutionalized adult glaucoma patients with initial trabeculectomies or drainage device implantations between April 4, 2014 and December 31, 2018. Patients were stratified into groups on the basis of straight-line distance from residence to University of North Carolina at Chapel Hill's Kittner Eye Center and distance from residence to interstate access. Corrective procedures, visual acuity, appointment cancellations, no-shows, and insurance data were recorded. Means were compared using 2-tailed Student t-test, Pearson χ, analysis of variance, and multivariate logistical regression determined odds ratios for loss to follow-up. RESULTS: In total, 199 patients met all inclusion criteria. Six-month postoperatively, patients >50 miles from clinic had greater odds of loss to follow-up compared with patients <25 miles (odds ratios, 3.47; 95% confidence interval, 1.24-4.12; P<0.05). Patients >50 miles from clinic had significantly more missed appointments than patients 25 to 50 miles away, and patients <25 miles away (P=0.008). Patients >20 miles from interstate access had greater loss to follow-up than those <10 miles (t(150)=2.05; P<0.05). Mean distance from clinic was 12.59 miles farther for patients lost to follow-up (t(197)=3.29; P<0.01). Patients with Medicaid coverage had more missed appointments than those with Medicare plans (t(144)=-2.193; P<0.05). CONCLUSIONS: Increased distance from clinic and interstate access are associated with increased missed appointments and loss to follow-up. Glaucoma specialists should consider these factors when choosing surgical interventions requiring frequent postoperative evaluations.

Association of Dry Eye Disease With Dyslipidemia and Statin Use.

PURPOSE: To determine whether an association exists between dry eye disease (DED) and statin use and/or dyslipidemia. DESIGN: Retrospective, case-control study. METHODS: Setting: University of North Carolina (UNC)-affiliated healthcare facilities. STUDY POPULATION: 72,931 patients seen at UNC ophthalmology clinics over a 10-year period. MAIN OUTCOME MEASURES: Odds ratios (ORs) calculated between DED and a history of low, moderate, or high-intensity statin use; and ORs calculated between DED and abnormal lipid panel values. RESULTS: Total of 39,336 individuals (53.9% female) were analyzed after exclusion of individuals with confounding risk factors for DED. Of these, 3,399 patients (8.6%) carried a diagnosis of DED. Low-, moderate-, and high-intensity statin regimens were used by 751 subjects (1.9%), 2,655 subjects (6.8%), and 1,036 subjects (2.6%). Lipid abnormalities were identified as total cholesterol >200 mg/dL, 4,558 subjects (11.6%); high-density lipoprotein (HDL) <40 mg/dL, 2,078 subjects (5.3%); low-density lipoprotein (LDL) >130 mg/dL, 2,756 subjects (7.0%); and triglycerides (TGs) >150 mg/dL, 2,881 subjects (7.3%). The odds ratios (OR) of carrying a diagnosis of DED given the presence of low-, moderate-, and high-intensity statin use were 1.39 (95% confidence interval [CI]: 1.13-1.72); OR 1.47 (95% CI: 1.30-1.65), and OR 1.46 (95% CI: 1.21-1.75), respectively. The OR of carrying a diagnosis of DED given the presence of total cholesterol >200 mg/dL, HDL <40 mg/dL, LDL >130 mg/dL, and TGs >150 mg/dL were 1.66 (95% CI: 1.52-1.82), 1.45 (95% CI: 1.26-1.67), 1.55 (95% CI: 1.39-1.74), and 1.43 (95% CI: 1.27-1.61), respectively. CONCLUSIONS: A history of statin use or dyslipidemia is associated with an increased odds of having a DED diagnosis. Further studies are needed to determine whether statin use and/or dyslipidemia increases the risk of DED.

Trends in open shoulder surgery among early career orthopedic surgeons: who is doing what?

BACKGROUND: The incidence of various open shoulder procedures has changed over time. In addition, various fellowships provide overlapping training in open shoulder surgery. There is a lack of information regarding the relationship between surgeon training and open shoulder procedure type and incidence in early career orthopedic surgeons. METHODS: The American Board of Orthopaedic Surgery Part-II database was queried from 2002 to 2016 for reported open shoulder procedures. The procedures were categorized as follows: arthroplasty, revision arthroplasty, open instability, trauma, and open rotator cuff. We evaluated procedure trends as well as their relationship to surgeon fellowship categorized by Sports, Shoulder/Elbow, Hand, Trauma, and "Other" fellowship as well as no fellowship training. We additionally evaluated complication data as it related to procedure, fellowship category, and volume. RESULTS: Over the 2002-2016 study period, there were increasing cases of arthroplasty, revision arthroplasty, and trauma (P < .001). There were decreasing cases in open instability and open rotator cuff (P < .001). Those with Sports training reported the largest overall share of open shoulder cases. Those with Shoulder/Elbow training reported an increasing overall share of arthroplasty cases and higher per candidate case numbers. The percentage of early career orthopedic surgeons reporting 5 or more arthroplasty cases was highest among Shoulder/Elbow candidates (P < .001). Across all procedures, those without fellowship training were least likely to report a complication (odds ratio [OR], 0.76; 95% confidence interval, 0.67-0.86; P < .001). Shoulder/Elbow candidates were least likely to report an arthroplasty complication (OR, 0.84, P = .03) as was any surgeon reporting 5 or more arthroplasty cases (OR, 0.81; 95% confidence interval, 0.70-0.94; P = .006). CONCLUSION: The type and incidence of open shoulder surgery procedures continues to change. Among early career surgeons, those with more specific shoulder training are now performing the majority of arthroplasty-related procedures, and early career volume inversely correlates with complications.

Predictors of referral to the North Carolina Infant-Toddler Program among infants with birth defects and infants born extremely preterm or low birthweight.

BACKGROUND: North Carolina's Infant-Toddler Program (NC ITP) provides Early Intervention (EI) services from birth to age three for children at risk for developmental delays. This study examined referral patterns by clinical and sociodemographic characteristics among infants with birth defects and infants born extremely preterm (gestational age < 27 weeks) or extremely low birthweight (<1,000 g). METHODS: A retrospective cohort of North Carolina resident births from 2012 to 2014 was matched to data from the North Carolina Birth Defects Monitoring Program and NC ITP records. A total of 2,463 infants with eligible birth defects and 2,118 extremely preterm or low birthweight infants were identified. Adjusted odds ratios and 95% confidence intervals from multivariable logistic regression models were used to analyze differences in referral by sociodemographic and clinical factors. Referrals resulting in enrollment were also examined. RESULTS: About 70% of infants with eligible birth defects and 85% of extremely premature infants were referred to the NC ITP. Geographic region, maternal race/ethnicity, maternal and infant enrollment in Medicaid, and hospital level of care at delivery were associated with referral among both at-risk groups. Among infants with birth defects, maternal age, education, and marital status were also associated with referral, as well as gestational age, birthweight, and the presence of multiple anomalies. Of the infants with referrals, over 80% in each group were subsequently enrolled. CONCLUSIONS: Many of the sociodemographic and clinical factors examined were associated with EI referral. These findings can be used to address coverage gaps and improve referral and enrollment rates for at-risk infants.

Racial and Ethnic Differences in the Association Between Diabetes Mellitus and Dry Eye Disease.

Purpose:To examine the association between dry eye and diabetes mellitus among a large North Carolina patient population. Methods:A retrospective cross-sectional study of patients seen within the University of North Carolina medical system between July 1, 2008, and September 1, 2017, was performed. De-identified medical records contained within the Carolina Data Warehouse of adult patients who have had an ocular evaluation were studied. Four categories of disease states were identified by ICD-9 and ICD-10 codes. Patients were grouped based on being diagnosed with Dry Eye Disease (ICD-9: 375.15, 370.33, and ICD-10: H04.12, H16.221), Diabetes Mellitus (ICD-9: 250.00-250.93 and ICD-10: E08-E11, E13), neither, or both diseases. Odds ratios of the association between diabetes and dry eye were calculated for the following racial/ethnic groups: Non-Hispanic White, Non-Hispanic Black/African American, Asian, and Hispanic. Results:A total of 81,480 patients were included in the analysis; of those, 8978 patients had dry eye disease and 18,361 patients had diabetes. The remaining 54,141 patients had neither disease. Dry eye prevalence among patients with diabetes was 14.39% (95% CI: 13.89-14.91%) and 10.11% (95%CI: 9.88-10.35%) among patients without diabetes. The odds of a patient with diabetes having dry eye are 1.15 (95% CI: 1.09-1.21) times that of a patient without diabetes. Asian patients with diabetes demonstrated the highest odds of having dry eye at 1.49 (95% CI: 1.12-1.98). Conclusions:Dry eye is common among patients with diabetes, and the association between these diseases is strongest among Asian patients. Race and ethnicity are important demographic factors that may guide providers in the diagnosis and treatment of dry eye in the setting of diabetes.

Association Between Dry Eye Disease and Migraine Headaches in a Large Population-Based Study.

Importance: Reports in the literature have conflicting findings about an association between dry eye disease (DED) and migraine headaches. Objective: To determine the strength of the association between DED and migraine headaches. Design, Setting, and Participants: This retrospective case-control study included 72 969 patients older than 18 years from University of North Carolina-affiliated health care facilities from May 1, 2008, through May 31, 2018. Deidentified aggregate patient data were queried; data were analyzed from June 1 through June 30, 2018. Exposures: Diagnosis of migraine headache. Main Outcomes and Measures: Odds ratios calculated between DED and migraine headaches for participants as a whole and stratified by sex and age group. Results: The base population consisted of 72 969 patients, including 41 764 men (57.2%) and 31 205 women (42.8%). Of these, 5352 patients (7.3%) carried a diagnosis of migraine headache, and 9638 (13.2%) carried a diagnosis of DED. The odds of having DED given a diagnosis of migraine headaches was 1.72 (95% CI, 1.60-1.85) times higher than that of patients without migraine headaches. After accounting for multiple confounding factors, the odds of having DED given a diagnosis of migraine headaches was 1.42 (95% CI, 1.20-1.68) times higher than that of patients without migraine headaches. Conclusions and Relevance: These findings suggest that patients with migraine headaches are more likely to have comorbid DED compared with the general population. Although this association may not reflect cause and effect if unidentified confounders account for the results, these data suggest that patients with migraine headaches may be at risk of carrying a comorbid diagnosis of DED.