SEX DIFFERENCES IN CARDIOVASCULAR RISK AND KIDNEY FUNCTION: CREATININE VS CYSTATIN C

SEX DIFFERENCES IN CARDIOVASCULAR RISK AND KIDNEY FUNCTION: CREATININE VS CYSTATIN C

Nicole De La Mata¹, Jennifer Lees², James Hedley¹, Michael Sullivan², Brenda Rosales¹, Patrick Mark², Angela Webster<sup>1,3,4

1</sup>Sydney School of Public Health, Faculty of Medicine and Health, University Of Sydney, Camperdown, NSW, Australia
²School of Cardiovascular and Metabolic Health, University of Glasgow, Glasgow, Glasgow, United Kingdom
³Centre for Renal and Transplant Research, Westmead Hospital, Westmead, NSW, Australia
⁴NHMRC Clinical Trials Centre, Sydney Medical School, Faculty of Medicine and Health, University of Sydney, Camperdown, NSW, Australia

Abstract

Aim:
We explored whether estimating kidney function using creatinine (eGFRcr), cystatin C (eGFRcys) or the combination of both (eGFRcrcys) improved prediction of cardiovascular risk by sex.

Background:
Incorporating cystatin C into cardiovascular risk prediction may offer better discrimination between sexes to identify higher risk individuals.

Methods:
We included all participants from the prospective research cohort UK Biobank without a prior cardiovascular event and who had kidney function measures available at baseline. We fitted adjusted cause-specific Cox proportional hazards models to test association with eGFRcr, eGFRcys and eGFRcrcys (mL/min/1.73m2) and interaction with sex for all-cause cardiovascular events, including arrhythmia, heart failure and myocardial infarction.

Results:
Among 432,015 eligible participants, 241,398 (56%) were women. Over 11.4 median years (IQR:10.5-12.1), there were 72,135 cardiovascular events including 31,961 arrhythmia, 28,268 myocardial infarction and 8,405 heart failure events. eGFRcys and eGFRcrcys were more strongly associated with increased cardiovascular risk in females than males (interaction term p<0.001), an association not seen with eGFRcr. Compared to eGFRcys >90-105, women with eGFRcys >30-45 had adjusted HR 2.46 (95%CI:2.20-2.76) for cardiovascular event risk, while males had adjusted HR 1.95 (95%CI: 1.75-2.18). Males and females had the same cardiovascular risk with eGFRcr, at adjusted HR 1.73 (95%CI:1.54-1.94) for >30-45 compared to >90-105. Concordance was greatest for adjusted model with eGFRcys (c=0.715).

Conclusion:
Evaluating kidney function using creatinine underestimates cardiovascular risk among women compared to cystatin C. Measurement of cystatin C improves estimation of cardiovascular risk associated with kidney function, particularly for women.

Biography

Dr Nicole De La Mata, BSc (Biology), MBiostats, PhD (epidemiology/biostatistics) is an early-mid-career biostatistician researcher and senior lecturer in biostatistics at the University of Sydney. Her expertise lies in establishing data linkage studies and using these to apply advanced statistical models to understand patient journeys and complex clinical scenarios. Her work has broadly aimed to improve health service quality and efficiency to translate into better and more equitable patient outcomes.