Hemodynamic and glucometabolic factors fail to predict renal function in a random population sample

Objective: To determine whether baseline hemodynamic and/or glucometabolic risk factors could predict renal function at follow-up, independently of baseline serum creatinine, in survivors from a random population sample. Design and method: We examined associations between baseline serum creatinine, hemodynamic factors (systolic blood pressure (SBP), heart rate (HR)), glucometabolic factors (fasting plasma glucose and insulin, 2-hour plasma glucose and insulin during oral glucose tolerance test (OGTT), a modified 40 minute oral disposition index (DIo), and Homeostatic Model Assessment (HOMA) derived indices of beta-cell function (HOMA-2B), insulin sensitivity (HOMA-2S), and insulin resistance (HOMA-2IR)), traditional cardiovascular risk factors (age, sex, smoking status, body mass index, diabetes mellitus, total serum cholesterol), and later renal function determined as serum cystatin C in 238 men and 7 women aged 38 to 49 years at the time of inclusion, using multivariable linear regression analysis (p-entry 0.05, p-removal 0.20). Study subjects came from a random population based sample and were included 1974-1992, whilst the follow-up with cystatin C measurement was performed 2002-2006. Results: Mean (+/- s.d.) follow-up time, creatinine, and cystatin C were 28 +/- 1 years, 94 +/- 12 micromoles/L, and 1.220 +/- 0.298 mg/L. Mean estimated glomerular filtration rates at baseline and follow-up were 84 +/- 12 mL/min/1.73m2 and 68 +/- 18 mL/min/1.73m2. In univariate analyses, cystatin C was positively associated with age, female sex, SBP, HOMA-2B, creatinine, and the time elapsed between inclusion and follow-up. In multivariable analysis, age (beta = 0.013 (95% confidence interval (CI), 0.002 to 0.024); p <0.02), female sex (beta = 0.271 (95% CI, 0.036 to 0.506); p = 0.02), creatinine (beta = 0.007 (95% CI, 0.004 to 0.010); p <0.001), and time span (beta = 0.062 (95% CI, 0.018 to 0.106); p = 0.006) remained significantly associated with cystatin C.We did not detect any significant interactions between SBP, HOMA-2B, and the other risk factors. Conclusions: Although SBP and HOMA-2B were both associated with later renal function as assessed by serum cystatin C, none of them remained significant in a multivariable model adjusted for age, sex, serum creatinine, and time from inclusion to follow-up.