Abstract
OBJECTIVE: Body mass index (BMI) inadequately captures heterogeneity in cardiovascular (CV) risk. We hypothesised that a 'Slow Burn' phenotype, defined as lower-than-expected energy intake relative to BMI, age, and sex, identifies individuals at elevated CV risk across the BMI spectrum.</p>
METHODS: We analysed 160 905 White/European participants from the UK Biobank. Relative caloric intake status was derived using residuals from a linear model predicting energy intake from BMI, age, and sex and categorised as low (LI; the 'Slow Burn' phenotype, < 30th percentile), normal (NI, 30th-70th), or high (HI, > 70th). Nine phenotypes were defined by combining intake groups with BMI categories: normal weight (NW), overweight (OW), and obese (OB). The primary outcome was 3-point major adverse cardiovascular events (3P-MACE: myocardial infarction [MI], stroke, and CV mortality).</p>
RESULTS: Compared with the NI group, the LI phenotype was associated with a higher risk of 3P-MACE (adjusted hazard ratio [aHR] 1.07, 95% CI 1.01-1.14) and MI (aHR 1.15, 95% CI 1.06-1.25), while the HI group showed no increased risk. In BMI-stratified analyses, the NW-LI (Normal-Weight Slow Burn) phenotype demonstrated an increased 3P-MACE risk (aHR 1.17; 95% CI 1.04-1.33) versus the NW-NI reference. Mediation analysis revealed that low muscle strength accounted for 5.2% (95% CI 1.6%-17.7%) of the excess risk in the normal-weight stratum, whereas MASLD showed no evidence of mediation. Longitudinally, NW-LI participants were more prone to future weight gain than their NI counterparts.</p>
CONCLUSION: Relative caloric intake identifies a high-risk 'Slow Burn' phenotype, characterised by increased cardiovascular risk despite lower-than-expected caloric intake. These findings highlight the limitations of BMI-based risk assessment and support risk-stratification strategies that incorporate energy balance and functional status.</p>