Differences between cystatin C- and creatinine-based estimated glomerular filtration rate and the risks of...

Estimated glomerular filtration rate (eGFR) based on cystatin C and creatinine often yields discordant results. Although the eGFR difference, defined as cystatin C-based eGFR minus creatinine-based eGFR, has been proposed as a marker of risk for adverse clinical outcomes, its association with postoperative complications and mortality remains unclear. This study aimed to investigate whether the eGFR difference is associated with postoperative complication and mortality risks.

Using data from the UK Biobank, the primary analysis included 26,065 participants with baseline serum creatinine and cystatin C levels who had surgery within 1 year of recruitment. Extended analyses included 241,006 participants who had major surgeries during the follow-up period (median follow-up, 13.7 years). Primary outcomes were 30-day postoperative complications and 90-day all-cause mortality. Sensitivity and subgroup analyses were performed.

In the primary analysis, restricted cubic spline modeling revealed an L-shaped association, with a threshold near -15 mL min^-1 1.73 m^-2. Compared with a neutral eGFR difference (≥ - 15 mL min^-1 1.73 m^-2), a highly negative eGFR difference (<-15 mL min^-1 1.73 m^-2) was associated with increased risks of complications (odds ratio [OR] = 1.69, 95% confidence interval [CI]: 1.38 to 2.08, P < 0.001) and mortality (OR = 2.29, 95% CI: 1.59 to 3.32, P < 0.001) in the unadjusted models. These associations remained significant after full adjustment (complications: OR = 1.60, 95% CI: 1.28 to 2.01, P < 0.001; mortality: OR = 1.96, 95% CI: 1.27 to 3.00, P = 0.002). Similar associations were observed in the extended analyses of participants having major surgery during follow-up (complications: OR = 1.23, 95% CI: 1.17 to 1.30, P < 0.001; mortality: OR = 1.50, 95% CI: 1.39 to 1.63, P < 0.001). Findings were consistent across sensitivity and subgroup analyses, with the associations attenuated for complications (OR = 1.36, 95% CI: 1.07-1.73, P = 0.01) and mortality (OR = 1.22, 95% CI: 0.79-1.89, P = 0.37) in fully adjusted models that included cystatin C-based instead of creatinine-based eGFR.

A highly negative eGFR difference was independently associated with increased risks of postoperative complications and mortality. The findings likely reflect a combination of more accurate assessment of eGFR by cystatin C as well as non-GFR determinants of circulating creatinine and cystatin C concentrations.