Search

Purpose: Electronic health records (EHRs) contain a vast amount of clinical data. Improved automated classification approaches have the potential to accurately and efficiently identify patient cohorts for research. We evaluated if a rule-based natural language processing (NLP) algorithm using clinical notes performed better for classifying proliferative diabetic retinopathy (PDR) and nonproliferative diabetic retinopathy (NPDR) severity compared with International Classification of Diseases, ninth edition (ICD-9) or 10th edition (ICD-10) codes. Design: Cross-sectional study. Subjects: Deidentified EHR data from an academic medical center identified 2366 patients aged ≥18 years, with diabetes mellitus, diabetic retinopathy (DR), and available clinical notes. Methods: From these 2366 patients, 306 random patients (100 training set, 206 test set) underwent chart review by ophthalmologists to establish the gold standard. International Classification of Diseases codes were extracted from the EHR. The notes algorithm identified positive mention of PDR and NPDR severity from clinical notes. Proliferative diabetic retinopathy and NPDR severity classification by ICD codes and the notes algorithm were compared with the gold standard. The entire DR cohort (N = 2366) was then classified as having presence (or absence) of PDR using ICD codes and the notes algorithm. Main Outcome Measures: Sensitivity, specificity, positive predictive value (PPV), negative predictive value, and F1 score for the notes algorithm compared with ICD codes using a gold standard of chart review. Results: For PDR classification of the test set patients, the notes algorithm performed better than ICD codes for all metrics. Specifically, the notes algorithm had significantly higher sensitivity (90.5% [95% confidence interval 85.7, 94.9] vs. 68.4% [60.4, 75.3]), but similar PPV (98.0% [95.4–100] vs. 94.7% [90.3, 98.3]) respectively. The F1 score was 0.941 [0.910, 0.966] for the notes algorithm compared with 0.794 [0.734, 0.842] for ICD codes. For PDR classification, ICD-10 codes performed better than ICD-9 codes (F1 score 0.836 [0.771, 0.878] vs. 0.596 [0.222, 0.692]). For NPDR severity classification, the notes algorithm performed similarly to ICD codes, but performance was limited by small sample size. Conclusions: The notes algorithm outperformed ICD codes for PDR classification. The findings demonstrate the significant potential of applying a rule-based NLP algorithm to clinical notes to increase the efficiency and accuracy of cohort selection for research. Financial Disclosure(s): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Abstract Background Intravitreal anti-vascular endothelial growth factor (anti-VEGF) injections are the standard of care for diabetic macular edema (DME), a common complication of diabetes. This study aimed to identify factors influencing DME intravitreal anti-VEGF treatment outcomes in real-world practice. Methods This was a multi-center retrospective observational study using medical chart review of participants receiving anti-VEGF injections for DME (N = 248). Demographic and clinical variables were assessed for association with best corrected visual acuity (BCVA) and central macular thickness (CMT) outcomes using regression models. Results There was a significant improvement in BCVA (p

Abstract Objectives To assess whether insulin therapy impacts the effectiveness of anti-vascular endothelial growth factor (anti-VEGF) injection for the treatment of diabetic macular edema (DME) in type 2 diabetes mellitus. Methods This was a retrospective multi-center analysis. The best-corrected visual acuity (BCVA) at 12 months, BCVA change, central macular thickness (CMT), CMT change, and cumulative injection number were compared between the insulin and the oral hypoglycemic agent (OHA) groups. Results The mean final BCVA and CMT improved in both the insulin (N = 137; p 0.05) and baseline HbA1c (p > 0.05). Conclusion Insulin therapy does not alter treatment outcomes for anti-VEGF therapy in DME.

Abstract Background Diabetic macular edema (DME) and proliferative diabetic retinopathy (PDR) are sight-threatening complications of diabetes mellitus and leading causes of adult-onset blindness worldwide. Genetic risk factors for diabetic retinopathy (DR) have been described previously, but have been difficult to replicate between studies, which have often used composite phenotypes and been conducted in different populations. This study aims to identify genetic risk factors for DME and PDR as separate complications in Australians of European descent with type 2 diabetes. Methods Caucasian Australians with type 2 diabetes were evaluated in a genome-wide association study (GWAS) to compare 270 DME cases and 176 PDR cases with 435 non-retinopathy controls. All participants were genotyped by SNP array and after data cleaning, cases were compared to controls using logistic regression adjusting for relevant covariates. Results The top ranked SNP for DME was rs1990145 (p = 4.10 × 10− 6, OR = 2.02 95%CI [1.50, 2.72]) on chromosome 2. The top-ranked SNP for PDR was rs918519 (p = 3.87 × 10− 6, OR = 0.35 95%CI [0.22, 0.54]) on chromosome 5. A trend towards association was also detected at two SNPs reported in the only other reported GWAS of DR in Caucasians; rs12267418 near MALRD1 (p = 0.008) in the DME cohort and rs16999051 in the diabetes gene PCSK2 (p = 0.007) in the PDR cohort. Conclusion This study has identified loci of interest for DME and PDR, two common ocular complications of diabetes. These findings require replication in other Caucasian cohorts with type 2 diabetes and larger cohorts will be required to identify genetic loci with statistical confidence. There is considerable overlap in the patient cohorts with each retinopathy subtype, complicating the search for genes that contribute to PDR and DME biology.

Purpose: To describe the appearance of the superficial and deep retinal capillary plexi in three patients with fovea plana of differing severity using spectral-domain optical coherence tomography angiography (OCTA). Observations: In the first case of grade 1 fovea plana (a patient with 20/25 vision), OCTA showed an orderly branching pattern of vessels from the superficial and deep retinal plexi extending to the center of the fovea. The second case of grade 3 fovea plana (20/30 vision) showed some disruption of the orderly vascular pattern with small caliber vessels from both superficial and deep layers densely covering the fovea center. Case 3 represented a patient with grade 4 fovea plana associated with PAX6 mutation and poor visual acuity. OCTA revealed a disorganized pattern of large and small caliber vessels from the superficial capillary network extending into the center of the fovea. Conclusions: and importance: Previously available imaging modalities were unable to specifically target different layers of the retinal vasculature. Using OCTA we have been able to show progressive changes in the vascular pattern in the deep and superficial retinal layers of patients with different grades of fovea plana. This novel imaging technique may play a role in the classification and assessment of patients with fovea plana.