Sex differences in cardiac risk and kidney function: serumcreatinine versus cystatin C

Abstract Background Cystatin C may better assess cardiovascular risk than serum creatinine for kidney function, but its accuracy may vary by sex. We evaluated sex differences in cardiac risk using estimated kidney function from either biomarker. Methods We included all adults from the UK Biobank without prior cardiac event who had kidney function and baseline data. We defined cardiac events and deaths using ICD-10 codes in hospital or death records. We fitted cause-specific Cox models to evaluate sex differences in cardiac outcomes using estimated glomerular filtration (mL/min/1.73m2) from serum creatinine (eGFRCr), cystatin C (eGFRCys) and both (eGFRCr-Cys). Results Among 394,920 adults (55% female), 19,689 (9%) females and 28,540 (16%) males had cardiac events. In adjusted models, eGFRCys and eGFRCr-Cys showed stronger associations with increased cardiovascular risk in females than when using eGFRCr (p < 0.001). Females with eGFRCys 45–59 had elevated cardiac risk (HR 1.08, 95% CI 1.03–1.14) compared to males with eGFRCys 90–104—an effect not seen with eGFRCr or eGFRCr-Cys. eGFRCr showed a J-shaped association with cardiac risk, being increased in males but reduced in females when eGFRCr ≥ 105 (females: HR 0.65, 95% CI 0.61–0.69; males: HR 1.18, 95% CI 1.12–1.24 versus males with eGFRCr 90–104). The risk of cardiac events was more linear in adjusted models with eGFRCys. Conclusions Measurement of cystatin C improves estimation of cardiac risk associated with kidney function, particularly for females. Incorporating eGFRCys, rather than eGFRCr, into cardiovascular risk assessment may be more important for early detection and management of high-risk females with CKD.