Author(s): KIM, Y.1, COLLINGS, P.J.1, WANG, M.1, CHEN, Z.1, SHI, Q.1, LUO, S.1, AU YEUNG, S.L.1, GONZALES, T.I.2, SHARP, S.J.2, BRAGE, S.2, Institution: THE UNIVERSITY OF HONG KONG, Country: HONG KONG, Abstract-ID: 2158

INTRODUCTION:
Cardiorespiratory fitness (CRF) is an important modifiable risk factor for atherosclerotic cardiovascular disease (ASCVD). It remains unknown, however, if the associations of CRF with ASCVD risk differ by genetic risk. We examined prospective relationships of CRF with incident ASCVD across genetic susceptibility strata.
METHODS:
This prospective study included 69,447 UK Biobank participants (mean age 58 years; 54% female) without prevalent ASCVD and who completed a sub-maximal cycle test. Maximal oxygen consumption (mlO2/min/kg) was estimated using a validated bike ergometry procedure, standardised within sex and age strata, and categorised to create low (20th percentile), intermediate (20-80th percentiles) and high (80th percentile) fitness groups. Genetic susceptibility to ischaemic heart disease was quantified with weighted polygenic risk scores, and was used to assign participants to low, intermediate, and high genetic risk groups. Cox regression models were used to estimate the hazard of developing ASCVD and its major sub-components. In line with international definitions, ASCVD included all documented cases of ischaemic heart disease (IHD: comprising myocardial infarction, angina, coronary revascularisation procedures), ischaemic stroke, and peripheral arterial disease. We investigated IHD and major ischaemic events (myocardial infarction and ischaemic stroke) as nested secondary outcomes.
RESULTS:
During a median follow-up of 12 years, there were 5,448 incident ASCVD, 4,177 IHD, and 2,299 major ischaemic events. Each 1-standard deviation difference in sex- and age-standardised CRF (equivalent to 5.6 mLO2/kg/min) was associated with an 8% (hazard ratio [95% confidence interval]: 0.92 [0.89-0.95]) lower hazard of ASCVD, independently of demographic, lifestyle and health-related factors, and genetic risk. There was weak evidence of multiplicative interaction (p=0.06) between CRF and genetic risk, suggesting that the associations might be slightly stronger in the low genetic risk group (0.87 [0.80-0.94]) compared with intermediate (0.94 [0.90-0.98] and high genetic risk groups (0.91 [0.86-0.97]). There was no evidence of additive interaction (p=0.78). The 10-year absolute risk differences between low and high CRF groups were 1.2% and 1.6% within high and low genetic risk categories, respectively. Similar patterns of associations were found for incident IHD and major ischaemic events.
CONCLUSION:
Across all strata of genetic risk for cardiovascular disease, higher CRF is strongly associated with lower risk of ASCVD and its major sub-components. Adults should be encouraged to increase or maintain CRF levels, irrespective of genetic susceptibility.