TABLE 2.
Longitudinal and cross-sectional effects of the GRSs on blood pressure traits (in mmHg per a 1-sd change in GRS), the longitudinal OR of adult hypertension between the extreme quintiles of the GRS and replications in Bogalusa data.
| GRS* | Longitudinal effect† from 1980 through 2007 |
Replication in Bogalusa: longitudinal effect |
Cross-sectional effect at baseline in 1980 |
Cross-sectional effect at 2007 |
Longitudinal OR‡ for hypertension in adulthood |
Replication in Bogalusa: OR for hypertension in adulthood |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| European ancestry | African-American ancestry |
||||||||||||
| Systolic BP$ (mmHg) |
Diastolic BP (mmHg) |
Systolic BP (mmHg) |
Diastolic BP (mmHg) |
Systolic BP (mmHg) |
Diastolic BP (mmHg) |
Systolic BP (mmHg) |
Diastolic BP (mmHg) |
Systolic BP (mmHg) |
Diastolic BP (mmHg) |
European ancestry |
African- American ancestry |
||
| 13-SNP|| combi nation GRS | β¶=0.47 p=0.008 | β=0.53 p=0.0003 | β=0.18 p=0.30 |
β=0.01 p=0.94 |
β=1.22 p=0.002 |
β=0.78 p=0.005 |
β=0.11 p=0.61 | β=0.35 p=0.10 | β=1.0 p=0.0006 | β=0.76 p=0.002 | OR=1.82, CI#=1.53–2.17, p<0.0001 | OR=0.95, CI=0.80–1.13, p=0.57 | OR=1.10, CI=0.87–1.40, p=0.44 |
| 5-SNP systolic BP GRS | β=0.21 p=0.23 | β=0.42 p=0.048 |
β=1.02 p=0.09 |
β=0.08 p=0.69 | β=0.08 p=0.70 | β=0.71 p=0.02 | β=0.38 p=0.12 | OR=1.40, CI=1.18–1.64, p=0.001 | OR=0.965, CI=0.78–1.20, p=0.75 | OR=0.889, CI=0.61–1.29, p=0.54 | |||
| 8-SNP diastolic BP GRS | β=0.50 p=0.0005 | β=0.10 p=0.60 |
β=1.76 p=0.002 |
β=0.07 p=0.74 | β=0.37 p=0.08 | β=0.78 p=0.01 | β=0.65 p=0.008 | OR=1.25, CI=1.06–1.49, p=0.03 | OR=0.94, CI=0.72–1.23, p=0.64 | OR=1.84, CI=1.13–3.02 p=0.02 | |||
GRS=genetic Risk Score.
Longitudinal models adjusted for measurement year, age (and age squared for systolic blood pressure), sex and BMI; cross-sectional and logistic models adjusted for age, sex, BMI and family history.
OR=Odds Ratio (between the extreme quintiles of the GRS).
BP=Blood Pressure. A substitution method was used for subjects under antihypertensive treatment in 2001 (N=58) or 2007 (N=143) by adding 15 mmHg to the systolic and 10 mmHg to the diastolic BP measured in 2001 or 2007.
SNP=Single Nucleotide Polymorphism. The SNPs in each GRS were as follows: rs16998073, rs1530440, rs3184504, rs1378942, rs16948048, rs9815354, rs11014166, rs2384550, rs17367504, rs11191548, rs12946454, rs38181 and rs2681492 in the 13-SNP combination GRS; rs17367504, rs11191548, rs12946454, rs381815 and rs2681492 in the 5-SNP systolic BP GRS and rs16998073, rs1530440, rs3184504, rs1378942, rs16948048, rs9815354, rs11014166 and rs2384550 in the 8-SNP diastolic BP GRS. SNPs rs3184504 and rs1378942 were not available for the African-American population of the Bogalusa Heart Study.
β=effect estimate (in mmHg per a 1-sd change in the GRS).
CI=95% Confidence Interval.