Akinori Hayashi1, Koji Takano2, Tsuguto Masaki2, Sonomi Yoshino2, Akifumi Ogawa2, Masayoshi Shichiri2. 1. Department of Endocrinology, Diabetes and Metabolism, Kitasato University School of Medicine, Kanagawa, Japan. Electronic address: ahayashi@kitasato-u.ac.jp. 2. Department of Endocrinology, Diabetes and Metabolism, Kitasato University School of Medicine, Kanagawa, Japan.
Abstract
AIMS: HbA1c and glycated albumin (GA) are used to monitor glycemia, but their accuracy to represent glycemic profiles in hemodialysis remains controversial. METHODS: Continuous glucose monitoring in 97 patients with type 2 diabetes (41 on hemodialysis [HD] and 56 without nephropathy) was analyzed to evaluate whether HbA1c and/or GA serve as appropriate glycemic profile markers. RESULTS: The average glucose significantly correlated with HbA1c in both HD group and group without nephropathy (r=0.59, P<0.0001; r=0.40, P<0.005). The slopes of linear regression lines were statistically indistinguishable (F=0.30, P=0.744), while the y-intercepts were significantly different (F=57.86, P<0.0001). GA showed strong correlation with the glycemic standard deviation (r=0.68, P<0.0001), and with the average glucose (r=0.42, P<0.001). Least square analysis revealed that only HbA1c, but not GA, was significantly associated with the average glucose (F=10.20, P<0.0005; F=0.38, P=0.5427), while only GA was significantly associated with the glycemic variability in HD group. CONCLUSIONS: In HD participants, HbA1c correlates with the average glucose more than GA, but underestimates it, and a correction formula of HbA1c can be developed as an appreciable marker. GA value itself reflects the average glucose, but less accurately than HbA1c, while it could serve as an indicator for hyperglycemia/hypoglycemia excursion.
AIMS: HbA1c and glycated albumin (GA) are used to monitor glycemia, but their accuracy to represent glycemic profiles in hemodialysis remains controversial. METHODS: Continuous glucose monitoring in 97 patients with type 2 diabetes (41 on hemodialysis [HD] and 56 without nephropathy) was analyzed to evaluate whether HbA1c and/or GA serve as appropriate glycemic profile markers. RESULTS: The average glucose significantly correlated with HbA1c in both HD group and group without nephropathy (r=0.59, P<0.0001; r=0.40, P<0.005). The slopes of linear regression lines were statistically indistinguishable (F=0.30, P=0.744), while the y-intercepts were significantly different (F=57.86, P<0.0001). GA showed strong correlation with the glycemic standard deviation (r=0.68, P<0.0001), and with the average glucose (r=0.42, P<0.001). Least square analysis revealed that only HbA1c, but not GA, was significantly associated with the average glucose (F=10.20, P<0.0005; F=0.38, P=0.5427), while only GA was significantly associated with the glycemic variability in HD group. CONCLUSIONS: In HDparticipants, HbA1c correlates with the average glucose more than GA, but underestimates it, and a correction formula of HbA1c can be developed as an appreciable marker. GA value itself reflects the average glucose, but less accurately than HbA1c, while it could serve as an indicator for hyperglycemia/hypoglycemia excursion.
Authors: Leila R Zelnick; Zona O Batacchi; Iram Ahmad; Ashveena Dighe; Randie R Little; Dace L Trence; Irl B Hirsch; Ian H de Boer Journal: Diabetes Care Date: 2020-08-11 Impact factor: 19.112