Photo Credit: Palmihelp
The following is a summary of “Use of Diagnostic Codes for Primary Open-Angle Glaucoma Polygenic Risk Score Construction in Electronic Health Record–Linked Biobanks,” published in the June 2024 issue of Ophthalmology by Tran et al.
Polygenic risk scores (PRSs) help predict the risk and outlook for glaucoma. Researchers conducted a retrospective study comparing how PRS works for primary open-angle glaucoma (POAG) when defined by International Classification of Diseases (ICD) codes vs. manual medical record review.
They studied POAG cases in Mount Sinai BioMe and Mass General Brigham (MGB) biobanks using ICD codes, which were confirmed with optical coherence tomography and visual fields. The absence of POAG was confirmed in a 5% sample using Intraocular pressure and cup disc ratio. Using genotype data from the U.K. Biobank records, a genome-wide POAG-PRS was calculated using the lassosum method. Also, the area of curves (AUC) was compared for POAG prediction based on ICD codes vs. medical records.
The result showed 804 BioMe and 367 MGB cases identified by ICD codes. In BioMe, positive predictive value was 53%, negative predictive value 96%, sensitivity 97%, and specificity 44%. In MGB, values were 55%, 97%, 97%, and 53%, respectively. The PRS AUCs for POAG using ICD codes vs. manual review in BioMe were not statistically different (P≥0.21), 0.77 vs. 0.75 for African ancestry, 0.80 vs. 0.80 for Hispanic, and 0.81 vs. 0.81 for European. Similar results were found in MGB (P≥0.18) 0.72 vs. 0.80 for Africans, 0.83 vs. 0.86 Hispanics, and 0.74 vs. 0.73 Europeans.
They concluded using ICD codes or manual review performed similarly for POAG-PRS in EHR-linked biobanks. While manual assessment may not always be needed, using ICD codes for glaucoma diagnosis requires caution due to the low specificity (44-53%).