Study design and patient selection
This multicenter, prospective, single-arm study was conducted at eleven tertiary hospitals in South Korea. A total of 323 adult patients aged less than 70 years with overt proteinuria (urinary protein to creatinine ratio [uPCR] ≥ 1 mg/mg) who visited the nephrology clinic between April 2009 and December 2011 were enrolled. Blood pressure was relatively well-controlled, ranging from 100/60 to 160/100 mmHg, in all patients. Low-salt diet was thoroughly educated in all patients at the time of enrollment. The exclusion criteria were as follows: an estimated glomerular filtration rate (eGFR) less than 30 mL/min/1.73m2, the use of non-interruptible drugs that affect proteinuria, patients who received immunosuppressive treatment during the preceding 6 months, severe proteinuria with uPCR > 10 mg/mg, serum albumin < 2.5 g/dL or refractory edema, uncontrolled DM with HbA1C > 9.0%, underlying conditions that may affect intrarenal RAS activity (e.g., renal artery stenosis, primary aldosteronism, or pheochromocytoma), cardiovascular events during the preceding 12 months (e.g., myocardial infarction, unstable angina, coronary artery bypass surgery, percutaneous transluminal coronary angioplasty, cerebrovascular accident, or transient ischemic attack), hypokalemia or hyperkalemia (K ≤ 3.5 or ≥ 5.5 mmol/L, respectively), abnormal liver function (aspartate transaminase or alanine transaminase higher than twice the upper limit of the normal range), allergic reaction to angiotensin-converting enzyme inhibitors (ACEi) or angiotensin II receptor blockers (ARBs), or pregnant or nursing patients.
Patients received only conventional supportive care during the pre-study period. If patients were already taking ACEi or ARB before the screening, the drugs were discontinued for at least 4 weeks. Valsartan treatment was started at the beginning of the study, and the drug dose was increased if patients presented with uPCR ≥ 1 mg/mg or blood pressure over 160/100 mmHg at 8 weeks after enrollment.
Sample collection and measurement of intrarenal RAS activity
Clinical data, including age, sex, underlying comorbidities, and blood pressure, were extracted from electronic medical records. Laboratory data, including hemoglobin, albumin, cholesterol, uric acid, blood urea nitrogen (BUN), serum creatinine (sCr), eGFR, electrolytes, and uPCR, were also collected. The eGFR was calculated based on the modified Modification of Diet in Renal Disease using sCr [19].
To measure urinary AGT and renin, overnight-fasting early-morning urine samples were collected before starting valsartan treatment. Samples were centrifuged at 2,500 rpm for 15 min at 4°C, and the supernatants were then stored at -70°C. The urine samples were thawed and centrifuged at 13,000 rpm for 2 min at 4°C just prior to use in the enzyme-linked immunosorbent assay (ELISA) for AGT or the radioimmunoassay for renin. Sandwich ELISAs were performed to quantify urinary AGT (uAGT) using mouse monoclonal and rabbit polyclonal antibodies against recombinant human AGT as previously reported [20, 21]. Highly purified human AGT (Calbiochem, Beeston, UK) was used as the standard. The correlation coefficient of the assay was greater than 0.99, and the detection limit of the ELISA system was 0.9 ng/mL. The urinary renin activity was measured using renin radioimmunoassay beads (TFB, Tokyo, Japan) and a Cobra-II gamma counter (Packard Bioscience, Meriden, CT, USA) as previously reported.[8] Prior to measurement, the urine sample was concentrated 5-fold with an Amicon Ultra-10 centrifugal filter device (Millipore, Cork, Ireland). The detection limit of the enzymatic-kinetic assay was 0.10 ng angiotensin I/mL/h. The renin concentration was determined from the angiotensin I-generating activity with 1 ng angiotensin I/mL/h corresponding to 2.6 pg human renin/mL [8, 22].
To correct for the difference in urinary concentrations, the urinary AGT and renin levels were divided by the urinary creatinine level, resulting in a urinary AGT/creatinine ratio (uAGT/Cr) and a urinary renin/creatinine ratio (uR/Cr), respectively. The uAGT/Cr and uR/Cr levels were then compared after natural logarithmic conversion due to skewed distributions. Delta ln(uAGT/Cr) [Δln(uAGT/Cr)] was defined as ln(uAGT/Cr) at 24 weeks minus the baseline ln(uAGT/Cr), and ΔuPCR was defined as uPCR at either 24 weeks or 5 years after enrollment minus the baseline uPCR.
Follow-up and outcome measures
The uPCR was measured at 8 weeks and 24 weeks after enrollment in all patients (Fig. 1). Baseline ln(uAGT/Cr) and ln(uR/Cr), as well as Δln(uAGT/Cr), were compared between patients with a uPCR decrement ≥ 1 mg/mg (decrement group) and patients with a uPCR decrement < 1 mg/mg (non-decrement group) during valsartan treatment. Furthermore, other possible prognostic factors for proteinuria decrement were analyzed using multivariable logistic regression analysis. Subgroup analyses regarding predictive factors for a uPCR less than 1 mg/mg and changes in eGFR were conducted for 56 patients who were followed up for 5 years after enrollment.
Statistical analyses
Continuous variables with a normal distribution are expressed as the mean ± standard deviation (SD), and variables with a non-normal distribution are expressed as the median with interquartile range (IQR). Group means were compared using the t-test and medians were compared using the Mann–Whitney test. Serial changes were compared using the paired t-test or the Wilcoxon signed-rank test, and repeated measures with analysis of variance (ANOVA), as appropriate. Pearson’s correlation analysis was used to test correlations between two continuous variables. Multivariable logistic regression analysis was performed using variables with P < 0.1 in the univariable models. For all tests, a P-value < 0.05 was considered statistically significant. All statistical analyses were conducted using IBM SPSS software, version 23 (IBM Corporation, Armonk, NY, USA).