Abstract:Aims: To investigate the quantitative importance of systolic blood pressure (SBP) for cardiovascular disease using a Mendelian randomization analysis and compare the risk associations with equivalent differences in standard measures of SBP after correction for regression dilution. Methods: In the China Kadoorie Biobank prospective study, 82,373 unrelated adults aged 30-79 years had SBP recorded and were genotyped for 384 candidate SNPs, including 32 SNPs previously associated with blood pressure. These SNPs were combined to generate a genetic score for SBP (GS-SBP) and estimate the associated relative risks (RR) and 95% confidence intervals (CI) of major cardiovascular events (stroke, and non-fatal MI or IHD death). The results were compared with the RRs for 20 mmHg higher SBP before and after correction for regression dilution (basal or usual SBP).The correlation coefficient between replicate measurements of SBP between baseline and a re-survey after 3 years was 0.6 and was used to correct for regression dilution. Results: The overall mean (SD) age at survey was 51 (10.7) years and mean (SD) SBP was 131.9 (22.2) mmHg, but only 4.8% reported use of any anti-hypertensive treatment. Overall, there was a 5 mmHg difference in the mean SBP between the top and bottom fifths of the GS-SBP. After adjustment for regression dilution, each 20 mmHg higher usual SBP was associated with RRs of 1.78, 2.45, 1.61, and 1.84 for ischaemic stroke, haemorrhagic stroke, non-fatal MI or IHD death and major cardiovascular events, respectively (Table). For GS-SBP, the corresponding RRs were much more extreme, being 2.48, 2.85, 1.85 and 2.69, respectively. Conclusions: This Mendelian randomization study demonstrates that the effects of long-term differences in SBP for risk of major cardiovascular events in this largely untreated population were almost 50% greater than those estimated for standard measures of SBP after correction for regression dilution.

Abstract 16532: Impact of Systolic Blood Pressure on Cardiovascular Disease in a Chinese Population: A Mendelian Randomization Study