Skip to main content
. 2018 Apr 30;8:6710. doi: 10.1038/s41598-018-25023-5

Table 4.

Association of uNAG excretion with G/A ratio in subjects with type 1 diabetes.

Variables Univariate P Multivariate P
Unadjusted β (95% CI) Adjusted β (95% CI)
uNAG, log unit 0.092 (0.018 to 0.166) 0.015 0.103 (0.016 to 0.189) 0.020
Age, year 0.001 (−0.058 to 0.060) 0.971 −0.001 (−0.005 to 0.004) 0.761
Women (vs. men) 0.089 (−0.040 to 0.220) 0.177 0.016 (−0.119 to 0.153) 0.811
BMI, kg/m2 −0.030 (−0.051 to −0.008) 0.005 −0.028 (−0.049 to −0.007) 0.007
Diabetes duration, year 0.105 (−0.024 to 0.234) 0.112 0.173 (0.041 to 0.313) 0.011
uACR, log unit 0.033 (−0.028 to 0.095) 0.286 −0.007 (−0.080 to 0.075) 0.952
Angiotensin blockers use (yes vs. no) −0.121 (−0.300 to 0.058) 0.185 −0.212 (−0.417 to 0.005) 0.055
FPG, mg/dL 0.014 (0.005 to 0.024) 0.002 0.016 (0.006 to 0.025) 0.002
Hemoglobin, g/dL −0.014 (−0.053 to 0.024) 0.471 0.006 (−0.033 to 0.045) 0.782

uNAG and uACR were log-transformed for linear regression models (log-uNAG and log-uACR). In multivariate model, β coefficient of uNAG for G/A ratio was reported with adjustment for age, sex, BMI, duration of diabetes, log-uACR, angiotensin blockers use, FPG, and hemoglobin. Bold characters represent statistically significant values. Abbreviations: uNAG, urinary N-acetyl-β-D-glucosaminidase; BMI, body mass index; uACR, urinary albumin-to-creatinine ratio.