1011 — Leveraging the EMR for Studying Genetic Determinants of Complex Disease: Novel Genetic Association with Longitudinal Blood Pressure
Bhatnagar V (VA San Diego Healthcare System), Liu L
(University of California, San Diego), O'Connor DT
(University of California, San Diego), Salem RM
(Broad Institute), Nievergelt C
(University of California, San Diego)
The electronic medical record (EMR) can be used to create virtual cohorts designed to study genetic markers of chronic diseases such as hypertension. In this study, we leveraged VA's EMR to develop a database to model the association between selected genotypes and mean arterial blood pressure (MAP) over an average eight-year period.
Volunteers (N = 1,527, mean age 62 years, 95% male) consented to participate in a pilot study, with consent to review medical records and consent to store and use DNA. Using the VISN 22 Data Warehouse, outpatient blood pressure data was collected over an average of eight years. Covariates and potential confounders, including antihypertensive medication use, were collected at baseline. Polymorphisms on several genes important in blood pressure regulation were selected for genotyping. Using a two-stage approach, an initial screen tested the association between genotypes and baseline MAP adjusting for age, and correcting for multiple comparisons and population stratification. A random effects model was then used to further explore the relationship between significant genotypes and repeated blood pressure measures in the three ancestral groups in this cohort (European Americans, African Americans, and Other).
The initial screen suggested that a promoter region polymorphism at PON1 C/T-108 (rs705379) was associated with baseline MAP. PON1 C/T-108 was also significantly associated with MAP data collected over approximately eight years (adjusted p = 0.0008); those with a TT genotype at PON1 C/T-108 had higher blood pressures across all three ancestral and age groups. While MAP generally decreased with age, MAP increased among middle age African Americans with a T-allele. PON1 C/T-108 was also associated with baseline MAP in an independent replication sample based on 857 African Americans with early hypertensive nephrosclerosis.
The EMR was used to uncover a novel genetic association with blood pressure. Results suggested an independent association between the T-allele at PON1 C/T -108 and MAP, with a possible interaction between age and genotype among African Americans.
The EMR can be leveraged to study genetic determinants of complex chronic disease, leading to new insights with respect to risk factors and health outcomes among Veterans. However, patient confidentiality and data security maintaining large genetic databases are important considerations.