Serum fructosamine and glycated albumin and risk of mortality and clinical outcomes in hemodialysis patients

Abstract

Objective: Assays for serum total glycated proteins (fructosamine) and the more specific glycated albumin may be useful indicators of hyperglycemia in dialysis patients, either as substitutes or adjuncts to standard markers such as hemoglobin A1c, as they are not affected by erythrocyte turnover. However, their relationship with long-term outcomes in dialysis patients is not well described.

Research design and methods: We measured fructosamine and glycated albumin in baseline samples from 503 incident hemodialysis participants of a national prospective cohort study, with enrollment from 1995-1998 and median follow-up of 3.5 years. Outcomes were all-cause and cardiovascular disease (CVD) mortality and morbidity (first CVD event and first sepsis hospitalization) analyzed using Cox regression adjusted for demographic and clinical characteristics, and comorbidities.

Results: Mean age was 58 years, 64% were white, 54% were male, and 57% had diabetes. There were 354 deaths (159 from CVD), 302 CVD events, and 118 sepsis hospitalizations over follow-up. Both fructosamine and glycated albumin were associated with all-cause mortality; adjusted HR per doubling of the biomarker was 1.96 (95% CI 1.38-2.79) for fructosamine and 1.40 (1.09-1.80) for glycated albumin. Both markers were also associated with CVD mortality [fructosamine 2.13 (1.28-3.54); glycated albumin 1.55 (1.09-2.21)]. Higher values of both markers were associated with trends toward a higher risk of hospitalization with sepsis [fructosamine 1.75 (1.01-3.02); glycated albumin 1.39 (0.94-2.06)].

Conclusions: Serum fructosamine and glycated albumin are risk factors for mortality and morbidity in hemodialysis patients.

Figure 1

Adjusted relative hazards of outcomes in 503 incident hemodialysis participants of the CHOICE study. A and B: Hazard of all-cause mortality with fructosamine and glycated albumin, respectively. C and D: Hazard of CVD mortality with fructosamine and glycated albumin, respectively. E and F: Hazard of first CVD event with fructosamine and glycated albumin, respectively. G and H: Hazard of first sepsis hospitalization with fructosamine and glycated albumin, respectively. Relative hazard predicted using Cox proportional hazards regression adjusted for demographic characteristics [age, race (white or other), sex, and educational status (completed high school or not)] and clinical and treatment factors [smoking history (ever smoked), systolic blood pressure, BMI, ICED score (0–3), CVD, albumin, hemoglobin, total cholesterol, and C-reactive protein]. Fructosamine and glycated albumin are modeled as restricted cubic splines with knots at the 10th, 50th, and 90th percentiles. The solid line is the adjusted HR of mortality; 10th percentile in the overall population was used as the reference (HR = 1). The dashed lines are the 95% CIs. Bars are the frequency histogram, showing the distribution of each serum marker; white bars represent those without diabetes and the gray bars represent those with diabetes.

Figure 1

Adjusted relative hazards of outcomes in 503 incident hemodialysis participants of the CHOICE study. A and B: Hazard of all-cause mortality with fructosamine and glycated albumin, respectively. C and D: Hazard of CVD mortality with fructosamine and glycated albumin, respectively. E and F: Hazard of first CVD event with fructosamine and glycated albumin, respectively. G and H: Hazard of first sepsis hospitalization with fructosamine and glycated albumin, respectively. Relative hazard predicted using Cox proportional hazards regression adjusted for demographic characteristics [age, race (white or other), sex, and educational status (completed high school or not)] and clinical and treatment factors [smoking history (ever smoked), systolic blood pressure, BMI, ICED score (0–3), CVD, albumin, hemoglobin, total cholesterol, and C-reactive protein]. Fructosamine and glycated albumin are modeled as restricted cubic splines with knots at the 10th, 50th, and 90th percentiles. The solid line is the adjusted HR of mortality; 10th percentile in the overall population was used as the reference (HR = 1). The dashed lines are the 95% CIs. Bars are the frequency histogram, showing the distribution of each serum marker; white bars represent those without diabetes and the gray bars represent those with diabetes.