The Difference Between Cystatin C- and Creatinine-Based Estimated GFR and Associations With Frailty and Adverse Outcomes: A Cohort Analysis of the Systolic Blood Pressure Intervention Trial (SPRINT).

RATIONALE &
OBJECTIVE: In prior research and in practice, the difference between estimated glomerular filtration rate (eGFR) calculated from cystatin C level and eGFR calculated from creatinine level has not been assessed for clinical significance and relevance. We evaluated whether these differences contain important information about frailty. STUDY
DESIGN: A cohort analysis of the Systolic Blood Pressure Intervention Trial (SPRINT). SETTING & PARTICIPANTS: 9,092 hypertensive SPRINT participants who had baseline measurements of serum creatinine, cystatin C, and frailty. EXPOSURE: eGFRs calculated using CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equations (eGFRcys and eGFRcr), and eGFRDiff, calculated as eGFRcys-eGFRcr. OUTCOMES: A validated 35-item frailty index that included questionnaire data for general and physical health, limitations of activities, pain, depression, sleep, energy level, self-care, and smoking status, as well as medical history, cognitive assessment, and laboratory data. We defined frailty as frailty index score>0.21 (range, 0-1). The incidence of injurious falls, hospitalizations, cardiovascular events, and mortality was also recorded. ANALYTICAL APPROACH: We used logistic regression to model the cross-sectional association of baseline eGFRDiff with frailty among all SPRINT participants. Adjusted proportional hazards regression was used to evaluate the association of eGFRDiff with adverse outcomes and mortality.
RESULTS: Mean age was 68±9 (SD) years, mean eGFRcys and eGFRcr were 73±23 and 72±20mL/min/1.73m2, and mean eGFRDiff was 0.5±15mL/min/1.73m2. In adjusted models, each 1-SD higher eGFRDiff was associated with 24% lower odds of prevalent frailty (OR, 0.76; 95% CI, 0.71-0.81), as well as with lower incidence rate of injurious falls (HR, 0.84; 95% CI, 0.77-0.92), hospitalization (HR, 0.91; 95% CI, 0.88-0.95), cardiovascular events (HR, 0.89; 95% CI, 0.81-0.97), and all-cause mortality (HR, 0.71; 95% CI, 0.63-0.82); P<0.01. LIMITATIONS: Gold-standard measure of kidney function and assessment of muscle mass were not available.
CONCLUSIONS: The difference between eGFRcys and eGFRcr is associated with frailty and health status. Positive eGFRDiff is strongly associated with lower risks for longitudinal adverse outcomes and mortality, even after adjusting for chronic kidney disease stage and baseline frailty.