We previously reported that urinary excretion rates of angiotensinogen (AGT) provide a specific index of the activity of the intrarenal renin-angiotensin system in angiotensin II-dependent hypertensive rats. Meanwhile, we have recently developed direct enzyme-linked immunosorbent assays (ELISAs) to measure plasma and urinary AGT in humans. This study was performed to test a hypothesis that urinary AGT levels are enhanced in chronic kidney disease (CKD) patients and correlated with some clinical parameters. Eighty patients with CKD (37 women and 43 men, from 18 to 94 years old) and seven healthy volunteers (two women and five men, from 27 to 43 years old) were included. Plasma AGT levels showed a normal distribution; however, urinary AGT-creatinine ratios (UAGT/UCre) deviated from the normal distribution. When a logarithmic transformation was executed, Log(UAGT/UCre) levels showed a normal distribution. Therefore, Log(UAGT/UCre) levels were used for further analyses. Log(UAGT/UCre) levels were not correlated with age, gender, height, body weight, body mass index, systolic blood pressure, diastolic blood pressure, serum sodium levels, serum potassium levels, urinary sodium-creatinine ratios, plasma renin activity, or plasma AGT levels. However, Log(UAGT/UCre) levels were significantly correlated positively with urinary albumin-creatinine ratios, fractional excretion of sodium, urinary protein-creatinine ratios, and serum creatinine, and correlated negatively with estimated glomerular filtration rate. Log(UAGT/ UCre) levels were significantly increased in CKD patients compared with control subjects (1.8801 ± 0.0885 vs. 0.9417 ± 0.1048; P = .0024). These data confirmed our earlier report and showed that a new ELISA assay is a valid approach for measuring urinary AGT.