Role of cardiorespiratory fitness in lifetime cardiovascular risk prediction models: a cross-sectional analysis

Abstract

BackgroundDespite well-established associations between cardiorespiratory fitness (CRF) and long-term cardiovascular disease risk, CRF remains underused in clinical practice. With a growing emphasis on longer-term risk, developments to UK cardiovascular disease risk equations such as the Joint British Societies' 3 (JBS3) calculator now see the prediction of both lifetime and short-term risk. However, unlike the US LifetimeRisk model, the JBS3 algorithm does not include CRF as a risk variable. The aim of this study was to examine whether differences in CRF values are reflected in long-term cardiovascular disease risk prediction scores using a risk equation that includes CRF in the design (LifetimeRisk) and one that does not (JBS3). MethodsA retrospective, cross-sectional analysis was performed on 81 male steelworkers who participated in the second phase of the Prosiect Sir Gâr workplace-based cardiovascular disease risk assessment initiative in Carmarthenshire, Wales. CRF was calculated with a validated and reliable submaximal protocol based on heart rate responses (Chester Step Test). Between-group CRF differences (excellent or good vs average or below average) were determined by independent sample t test or Mann-Whitney U test, and any association between CRF and individual risk scores investigated by Pearson's correlation. FindingsMen in the lower CRF categories had greater body-mass index scores than men in the higher CRF categories (p=0·001); and less desirable blood lipid profiles with higher total cholesterol (p=0·041) and non-HDL cholesterol (p=0·019) concentrations. The predicted long-term cardiovascular disease risk in the LifetimeRisk model was also greater in the lower fitness group than higher fitness group (mean 13·6% [SD 2·9] vs 7·2 [1·9], p<0·0001), and a negative correlation was evident between individual CRF values and predicted risk scores (r=–0·603, p<0·0001). However, there were no differences between CRF groups in predicted JBS3 lifetime risk (median 79 years [IQR 77–80] vs 79 [78–80], p=0·354). InterpretationThese preliminary observations demonstrate that the LifetimeRisk prediction scores more accurately represent the current literature between CRF and longer-term cardiovascular disease risk in a group of white men. Although, more research is required, this study raises an important question as to whether CRF should be considered in the design of future UK long-term cardiovascular disease risk equations. FundingThis work was part funded by the European Social Fund through the European Union's Convergence programme administered by the Welsh Government.