Elevated urinary N-acetyl-β-D-glucosaminidase is associated with high glycoalbumin-to-hemoglobin A1c ratio in type 1 diabetes patients with early diabetic kidney disease

Abstract

Urinary N-acetyl-β-D-glucosaminidase (uNAG) predicted the progression of diabetic kidney disease (DKD) prior to development of albuminuria in diabetes patients. We sought whether uNAG level is associated with glycoalbumin-to-hemoglobin A1c ratio (G/A ratio), a marker of postprandial hyperglycemia and glycemic excursion, independent of albuminuria and kidney function. The association between uNAG excretion and G/A ratio was assessed in 204 consecutive subjects with type 1 diabetes (T1D) (mean age 43.9 years; 49.0% men). uNAG excretion level increased along with older age, hyperglycemia, and degree of albuminuria, but was not correlated with body mass index or estimated glomerular filtration rate (eGFR). Elevated uNAG showed robust association with higher G/A ratio (adjusted β = 0.103, P = 0.020) after adjustment for age, sex, body mass index, duration of diabetes, uACR, angiotensin blockers use, fasting plasma glucose, and hemoglobin level. uNAG showed better discriminatory performance for individuals with high G/A ratio than albuminuria (AUC 0.613 vs. 0.518, P = 0.038). Measurement of uNAG improved AUC for high G/A ratio from 0.699 to 0.748 (P = 0.043) when added to conventional risk factors (cutoff 5.24 U/g creatinine; sensitivity 62.5% and specificity 58.0%). In conclusion, Elevated uNAG was found to be associated with high G/A ratio in patients with T1D with early stage DKD, independent of age and albuminuria.

Figure 1

Comparison of uNAG level by combining the glycoalbumin-to-hemoglobin A1c (G/A) ratio and presence of albuminuria in subjects with type 1 diabetes. Low G/A ratios are indicated by values less than 2.8, whereas G/A ratios equal to 2.8 or more were defined as high. Albuminuria was defined as urinary albumin-to-creatinine ratio equal to or greater than 30 mg/g creatinine. P values for comparison between groups were calculated using Dunn’s procedure in order to correct for multiple comparison error. Abbreviations: uNAG, urinary N-acetyl-β-D-glucosaminidase; G/A ratio, glycoalbumin-to-hemoglobin A1c ratio.

Figure 2

Measurement of uNAG improved the identification of subjects with higher G/A ratios, in addition to conventional risk factors such as uACR. The solid line, very short dash line, dash line, and line with small cross represent the discriminatory ability characterized by the area under the curve (AUC) for log-uNAG alone, log-uACR alone, for the model that included conventional risk factors (log-uACR, age, sex, BMI, diabetes duration, angiotensin blocker use, eGFR, and fasting plasma glucose level), and for the composite of log-uNAG and the conventional risk model, respectively. Abbreviations: log-uNAG, log-transformed urinary N-acetyl-β-D-glucosaminidase; log-uACR, log-transformed urinary albumin-to-creatinine ratio; ROC, receiver operating characteristics curve; BMI, body mass index; eGFR, estimated glomerular filtration rate; G/A ratio, glycoalbumin-to-hemoglobin A1c ratio.