Mid-adulthood risk factor profiles for CKD

Abstract

Early identification of CKD risk factors may allow risk factor modification and prevention of CKD progression. We investigated the hypothesis that risk factors are present ≥30 years before the diagnosis of CKD in a case-control study using data from the Framingham Offspring Study. Patients with incident CKD (eGFR≤60 ml/min per 1.73 m2) at examination cycles 6, 7, and 8 were age- and sex-matched 1:2 to patients without CKD at baseline (examination 5). CKD risk factors were measured at each examination cycle. Logistic regression models, adjusted for age, sex, and time period, were constructed to compare risk factor profiles at each time point between cases and controls. During follow-up, 441 new cases of CKD were identified and matched to 882 controls (mean age 69.2 years, 52.4% women). Those who ultimately developed CKD were more likely to have hypertension (odds ratio [OR], 1.76; 95% confidence interval [CI], 1.23 to 2.51), obesity (OR, 1.71; 95% CI, 1.14 to 2.59), and higher triglyceride levels (OR, 1.43; 95% CI, 1.12 to 1.83) 30 years before CKD diagnosis, and were more likely to have hypertension (OR, 1.38; 95% CI, 1.07 to 1.79), higher triglyceride levels (OR, 1.35; 95% CI, 1.11 to 1.64), lower HDLc (OR, 0.89; 95% CI, 0.81 to 0.97), and diabetes (OR, 2.90; 95% CI, 1.59 to 5.29) 20 years before CKD diagnosis. These findings demonstrate that risk factors for CKD are identifiable ≥30 years before diagnosis and suggest the importance of early risk factor identification in patients at risk for CKD.

Keywords: CKD; clinical epidemiology; obesity; risk factors.

Figure 1.

Differences in blood pressure and hypertension in cases and controls up to 30 years before CKD diagnosis. Comparison of the mean systolic BP (A), diastolic BP (B), and proportion of participants with hypertension (C) in cases and controls at the time of CKD diagnosis and at time points 10, 20, and 30 years before diagnosis. The P values and SEM bars are derived from the logistic regression model.

Figure 2.

Risk factor profiles in cases and controls up to 30 years prior to CKD diagnosis. Comparison of the log triglycerides (A), mean HDL_c (B), and proportion of participants with diabetes (C) in cases and controls at the time of CKD diagnosis and at time points 10, 20, and 30 years before diagnosis. The P values and SEM bars are derived from the logistic regression model.

Figure 3.

Measures of body mass and obesity in cases and controls up to 30 years prior to CKD diagnosis. Comparison of the mean BMI (A) and the proportion of participants with obesity (B) in cases and controls at the time of CKD diagnosis and at time points 10, 20, and 30 years before diagnosis. The P values and SEM bars are derived from the logistic regression model.

Figure 4.

OR of CKD by number of risk factors at time points 10, 20, and 30 years before CKD diagnosis. Risk factors include diabetes, hypertension, smoking, and obesity and the analysis is adjusted for age, sex, and calendar year of CKD diagnosis. *P<0.05 (significant difference from referent values; 0 risk factors).