Comparative Effects of Antihypertensives on Proteinuria: Angiotensin-Converting Enzyme Inhibitor Versus a.-Antagonist Mark E. Rosenberg, MD, and Thomas H. Hostetter, MD • Control of hypertension improves the course of renal disease. We compared the renal hemodynamic and permselective responses to an angiotensin-converting enzyme inhibitor (CEI) (enalapril) and an /l!1-antagonist (prazosin) in 14 patients with established glomerular disease. A single-blinded, randomized, cross-over design was used consisting of a 3-week baseline period followed by two 4-week treatment periods, which were separated by a 4-week washout period. During the treatment periods, the CEI or /l!1-antagonist was added to the patients' baseline antihypertensive medications. Mean arterial pressure (MAP) was reduced to similar levels by both drugs, although the time-averaged blood pressure throughout the study was higher with the /l!1-antagonist. Twenty-four-hour urinary protein, albumin, and IgG excretion were not significantly different at the end of the CEI and /l!1-antagonist periods. However, compared with baseline values, significant decreases in total protein and IgG excretion occurred only during the CEI period, while albumin excretion decreased with both drugs. A 22% decrease in the fractional clearance of albumin (4.95 ± 1.44 to 3.88 ± 1.57 x 10-3 ; P < 0.01) and a 49% decrease in the fractional clearance of IgG (1.58 ± 0.42 to 0.81 ± 0.28 X 10-3 ; P < 0.001) occurred during CEI therapy with no significant changes in these paramaters being seen with /l!1-antagonist therapy (albumin: 4.95 ± 1.44 to 4.48 ± 1.51 X 10-3 , P = NS; IgG: 1.58 ± 0.42 to 1.71 ± 0.70 X 10-3 , P = NS). At the time of the fractional clearance measurements, MAP proved to be lower on the CEI. Reanalysis of the data for the subgroup of 11 patients without differences in MAP during the clearance period demonstrated a beneficial effect favoring CEI. Except for the greatest decreases in blood pressure (21 to 30 mm Hg), a greater antiproteinuric effect for a given decrease in blood pressure was seen with the CEI. Additionally, reduction in proteinuria occurred in a subset of seven patients whose baseline MAP was in the normotensive range. In conclusion, lowering MAP improves proteinuria. CEI appears to exert a more favourable effect even at similar MAP. Reductions in blood pressure, even within the accepted normal range, lessen permselective defects. © 1991 by the National Kidney Foundation, Inc. INDEX WORDS: Antihypertensive; proteinuria; converting enzyme inhibitor; renin; angiotensin; enalapril; prazosin.
I
N PATIENTS with chronic renal disease, control of hypertension reduces the rate of decay in renal function. I-3 Although a variety of antihypertensive agents have this beneficial effect, experimental evidence suggests that angiotensinconverting enzyme inhibitors (CEI) have a more specific effect on reducing renal injury, which may be related to their ability to lower glomerular capillary pressure. 4-9 Although numerous studies have examined the effects ofCEI in human renal disease,9-24 few studies have tested for a differential effect between CEI and other antihypertensives in humans. 17-24 The most definitive evidence for a beneficial effect of a particular antihypertensive on the progression of renal dis-
ease is dependent on demonstrating a decrease in the decline of glomerular filtration rate (GFR) and in progression to end stage. However, quantitating changes in permselectivity provides a useful shorter-term prognostic marker. 25-27 The purpose ofthe current study was to compare rigorously and directly the effects of two antihypertensives on renal function. Using a single-blinded, randomized, cross-over design, we compared the effects of an angiotensin-CEI (enalapril) and an lXI-antagonist (prazosin) on renal hemodynamics and glomerular permselectivity in patients with a variety of glomerular diseases. METHODS
From the Department ofMedicine, University ofMinnesota, Minneapolis, MN. Supported by National Institutes ofHealth Grants No. AM 31437 (TH.H.) and 5-M01-RR00400 (General Clinical Research Center), and by a grant from Merck Sharp and Dohme Research Laboratories. M.E.R. was supported by a research fellowship from the Medical Research Council of Canada. Address reprint requests to Mark E. Rosenberg, MD, University of Minnesota, Department of Medicine, Box 736 UMHC, 516 Delaware St SE, Minneapolis, MN 55455. © 1991 by the National Kidney Foundation, Inc. 0272-6386/91/1804-0007$3.00;0 472
Patients Fourteen patients were studied, nine with biopsy-proven primary glomerular disease, four with diabetic nephropathy as defined by proteinuric renal disease in the setting of longstanding diabetes mellitus and underlying retinopathy, and one who did not have a renal biopsy but clinically had primary renal disease (Table I). All patients were being treated for hypertension. Their antihypertensive medications, baseline mean arterial pressure (MAP), which was the average of four weekly readings, baseline creatinine clearances, and baseline proteinuria are shown in Table 1. Of note, seven patients had a MAP < 107 mm Hg, corresponding to a blood pressure of
American Journal of Kidney Diseases, Vol XVIII, No 4 (October), 1991: pp 472-482
COMPARATIVE EFFECTS OF ANTIHYPERTENSIVES
473
Table 1. Patient Characteristics
No.
Disease
Sex/Age
M/29
HTN
2 3 4
F/35 M/39 M/23
Diabetes Focal sclerosis Diabetes
5 6 7 8 9 10 11 12 13 14
M/29 F/32 F/46 F/39 M/65 M/62 M/46 F/30 F/31 M/34
Focal sclerosis Diabetes Membranous IgA nephropathy Membranous Diabetes Membranous SLE Focal sclerosis No biopsy
Antihypertensive Medications
Furosemide, metoprolol, minoxidil Furosemide Metoprolol Furosemide, metoprolol, hydralazine Bumetanide Furosemide Atenolol Furosemide, atenolol Furosemide Metolazone, clonidine, nifedipine Furosemide, verapamil Furosemide, propranolol Furosemide, verapamil Bumetanide
Baseline MAP (mmHg)
Baseline CrCI (mL/s/m~ [mL/min/ 1.73 m2l
Baseline Proteinuria (g/24h)
112
0.17 [18]
3.6
108 114 117
0.27 [28] 0.19 [20] 0.40 [42]
7.9 7.5 6.4
120 100 102 90 111 106 118 97 98 94
0.13 [14] 0.45 [47] 0.40 [42] 0.20 [21] 0.83 [86] 0.19 [20] 0.46 [48] 0.50 [52] 0.37 [38] 0.13[14]
5.2 0.9 1.2 0.7 8.5 13.1 3.6 2.5 4.6 2.4
Abbreviations: HTN, hypertension; SLE, systemic lupus erythematosus.
less than 140/90 and would, therefore, be considered to have adequate blood pressure control. 28 Entry criteria for the study were age between 18 and 65 years, serum creatinine between 133 and 442 JLmolfL/1.73 m 2 (1.5 and 5 mg/dL/1.73 m 2), dipstick-positive proteinuria, and no obvious reversible causes of their renal disease. If the patients were taking either of the two study medications, enalapril or prazosin, these were discontinued or substituted at least 4 weeks before the start of the study. Informed consent was obtained from all patients prior to entry into the study. The study was approved by the committee on human experimentation.
Design A single-blind, randomized, cross-over design was used. The total study duration was 15 weeks and consisted of a 3-week baseline period and two 4-week treatment periods, which were separated by a 4-week washout period. Throughout the study, the patients were seen at weekly intervals in the General Clinical Research Center (GCRC) at the University of Minnesota. The study began with a 3-week baseline period during which the patients were maintained on their previous antihypertensives (Table 1) and was followed by the first 4-week treatment period during which either the CEI (enalapriI) or the aI-antagonist (prazosin) was added to their previous antihypertensive medications. Following the first treatment period, the patients returned to their baseline medications for 4 weeks (washout period) before entering the second 4-week treatment period during which the second drug was added to their baseline medications. The order of the treatment periods was chosen by random assignment. Eight patients began with the CEI and six patients with the aI-antagonist. Both drugs were prepared in identical gelatin capsules insuring the single-blindedness of the study. The starting dose of the CEI was 2.5 mg
orally twice daily and of the aI-antagonist 2 mg orally twice daily. The goal of the treatment periods was to lower the MAP 10 mm Hg compared with baseline values. To insure equivalent reductions in blood pressure, the dosages of both drugs were adjusted on a weekly basis. On the night before the beginning and end of each treatment period, the patients were admitted to the GCRC. The following morning, blood was obtained in the supine position for measurement of plasma renin activity (PRA) and angiotensin-converting enzyme (ACE) activity. Measurement of PRA was repeated after I hour of standing. Renal clearance studies were then performed as outlined below. In all patients, weekly measurements of 24-hour urinary sodium, potassium, urea nitrogen, creatinine, and total protein; and serum sodium, potassium, chloride, bicarbonate, blood urea nitrogen (BUN), and creatinine were performed during all baseline and treatment visits. Hemoglobin, white blood cell (WBC) count, and 24-hour urinary albumin and IgG were measured every other week. Baseline values are those obtained before addition of the study drugs.
Blood Pressure Measurement Blood pressure was measured in an identical fashion throughout the study by a group of nurses who were blinded as to the study drug. Two blood pressure measurements 5 minutes apart were made in the supine position after 10 minutes of rest. Two standing blood pressure measurements 5 minutes apart were than performed. MAP was calculated by adding one third of the pulse pressure to the diastolic pressure. The mean value for MAP of all four readings was used to adjust the dose of the study drugs.
Clearance Studies Clearance studies were performed between 8:00 AM and noon at the beginning and end of each treatment period as
474
ROSENBERG AND HOSTETTER
Table 2. General Parameters
Weight (kg) Sodium (mmoI/L) Potassium (mmol/L) Urea nitrogen (mmol/L) [mg/dL] Creatinine (ILmoI/L) [mg/dL] CrCI (mL/s/m 2 ) [mL/min/1.73 m2] Total protein (gIL) [mg/dL] Albumin (gIL) [mg/dL] IgG (gIL [mg/dL] Hemoglobin (gIL) [g/dL]
Baseline
GEl
a,-Antagonist
75 ± 4 139 ± 1 4.2 ± 0 .2 16.0 ± 1.4 [45 ± 4] 283 ± 44 [3.2 ± 0.5] 0.35 ± 0.05 [36 ± 5] 58 ± 2 [5.8 ± 0 2] . 31 ± 2 [3.1 ± 0.2) 8.6 ± 0.8 [858 ± 76] 119±3 [11.9±0.3]
74 ± 4 138 ± 1 4.6 ± 0.1 18.6 ± 2.1 [52 ± 6] 301 ± 44 [3.4 ± 0.5] 0.33 ± 0.06 [34 ± 6] 59 ± 2 [5.9 ± 0.2] 30 ± 2 [3.0 ± 0.2] 8.6 ± 0.8 [856 ± 78] 116 ± 4 [11.6 ± 0.4]
76 ± 4 139 ± 1 4.1 ± 0.1 16.4±1 .4 [46±4] 292 ± 44 [3.3 ± 0.5] 0.37 ± 0.06 [38 ± 6] 58 ± 2 [5.8 ± 0.2] 31 ± 2 [3.1 ± 0.20) 8.5 ± 0.6 [853 ± 64] 117±4 [11 .7 ± 0.4]
NOTE. Values are mean ± SEM. No significant differences were detected.
described previously.29 Two intravenous catheters were inserted, one for the infusion of inulin (lso-Tex Diagnostics, Friendswood, TX) and p-aminohippurate (PAH) (Merck Sharpe & Dohme, West Point, PAl and the other in the opposite arm for repeat blood sampling. Throughout the clearance studies, the patients drank 250 mL of tap water every 30 minutes to maintain adequate urine flow. A loading dose of inulin and PAH was infused over 10 minutes followed by a continuous infusion of inulin and PAH designed to maintain blood levels of25 mgjdL and 2 mgjdL, respectively. Following a 60-minute equilibration period, three 30-minute urine collections were performed. Blood pressure and a blood sample for measurement of inulin and PAH were obtained at the midpoint of each clearance period. During the second clearance period, blood and urine were also obtained for measurement of albumin and IgG to enable calculation of the fractional clearances of these proteins. The fractional clearances were calculated by dividing the clearance of the protein by the clearance of the filtration marker inulin. The reabsorption of tl2-microglobulin was calculated by subtracting the urinary excretion of tl2-microglobulin from its filtered load. The average value for inulin and PAH clearances during the three periods was considered to represent GFR and renal plasma flow (RPF), respectively. PAH clearance was corrected for published values of extraction ratio for patients with renal disease by dividing the PAH clearance by 0.7. 30
Laboratory Measurements Urine and heparinized plasma were used for colorimetric measurements of inulin and P AH. 31 .32 Albumin and IgG were measured by rate nephelometry (Beckman Immunochemistry System; Beckman Instruments, Brea, CAl. Total protein was measured by the biuret technique. An enzyme immunoassay (Phadezym tlrmicroTest; Pharmacia Diagnostics, Piscataway, NJ) was used to determine tlrmicroglobulin. Serum sodium, potassium, chloride, bicarbonate, glucose, calcium, urea nitrogen, creatinine, and urine sodium, potassium, urea nitrogen, and creatinine were measured by an autoanalyzer (Beckman ASTRA). Hemoglobin and WBC count were determined by coulter counter. PRA was determined by radioimmunoassay for angiotensin I at pH 5.7 using a commercially available kit (DuPont Diagnostic Imaging, North Billerica, MA). Serum ACE activity
was measured using the ACE Microvial Radioassay system (Ventrex Laboratories, Portland, ME).
Statistics All data are reported as the mean ± SEM. Statistical differences between baseline and treatment values and between CEI and ai-antagonist values were calculated by either the Student's paired t test or the Wilcoxon signed rank test depending on whether the values were normally distributed or not. A statistically significant difference between values was considered for a P value less than 0.05.
RESULTS
General Parameters
The average daily dose (mg/d) of the CEI was week 1,5.0 ± 0.0; week 2,5.7 ± 0.6; week 3, 6.6 ± 0.9; week 4, 7.0 ± l.0; and of the aI-antagonist, week 1,3.9 ± 0.1; week 2,5.1 ± 0.5; week 3, 6.6 ± 1.0; week 4, 7.6 ± 104. Hyperkalemia requiring potassium-binder therapy occurred transiently in one patient on the CEI and two patients on the ai-antagonist. Other side effects included dyspepsia in one patient while taking the CEI and orthostatic hypotension, and nasal stuffiness each occurring in one patient on aI-antagonist therapy. No significant differences in serum sodium, potassium, BUN, creatinine, total protein, albumin, IgG, hemoglobin, creatinine clearance, or weight were seen throughout the study, although serum potassium tended to be higher with the CEI (Table 2). Serum chloride, bicarbonate, blood sugar, and WBC count were also similar throughout the study (data not shown). ACE activity was significantly lower and supine and standing PRA significantly higher at the end of the CEI period compared with both baseline and aI-antagonist values, establishing the patients'
475
COMPARATIVE EFFECTS OF ANTIHYPERTENSIVES Table 3. Angiotensin-Converting Enzyme and Plasma Renin Activity
Hemodynamics
Consistent with the goals of the study, both supine and standing MAP were significantly reduced by both drugs compared with baseline MAP except for the supine day 14 MAP of the ai-antagonist, when a numerical but not a statistical difference was seen (Fig 2). At the end of the treatment periods, there was no difference between the two drugs for either supine MAP (CEI 95 ± 4 val-antagonist 99 ± 3 mm Hg; P = NS) or standing MAP (CEI 94 ± 4 val-antagonist 95 ± 4 mm Hg; P = NS). As an estimate of the time-averaged MAP throughout the study, we added the individual MAP readings from day 7, 14,21, and 27 and divided by 4. This estimate of time-averaged MAP was lower with the CEI compared with the ai-antagonist (94 ± 3 v 98 ± 3 mm Hg; P < 0.05). As well, during the clearance period MAP was lower with CEI compared with ai-antagonist (CEI 101 ± 4 val-antagonist 108 ± 4 mm Hg; P < 0.05) despite administration of both drugs on the morning of the study and identical clearance protocols. No significant differences were seen for GFR, RPF, or filtration fraction between the CEI and ai-antagonist. Compared with baseline values, a significant decrease in GFR occurred with both drugs (Table 4). No correlation was present between changes in GFR and the decrease in blood pressure.
PRA (ng AI/mL/h) ACE (U/mL)
Supine
Standing
Baseline 4.51 ± 0.41 1.92 ± 0.33 5.41 ± 1.97 CEI 0.20 ± 0.02*t 9.05 ± 3.08*t 12.71 ± 4.00*t al-antagonist4.41 ± 0.37 1.79 ± 0.34 4.13 ± 1.14 NOTE. Values are mean ± SEM. P < 0.05 v baseline. t P < 0.05 val-antagonist. *
compliance with CEI (Table 3). Urinary excretion of sodium and urinary urea nitrogen were similar throughout the study, indicating constancy of dietary sodium and protein intake (Fig 1). Using all of the values for 24-hour urinary sodium measured throughout the study, the range of sodium excretion was 48 to 435 mmol/d, with a median of 154 mmol/d, and a mean ± SE of 166 ± 5 mmol/d. Similarly, the range of 24-hour urinary urea excretion ranged between 160 to 630 mmol/d (4.48 to 17.65 g nitrogen [N]/d), with a median of296 mmol/d (8.29 g N/d), and a mean ± SE of 323 ± 9 mmol/d (9.04 ± 0.24 g N/d). Assuming the patients were in steady-state protein balance, estimated protein intakes were calculated based on the following formula: Protein Intake
=
[UUN (g N/24 h)
X [0.031 X weight (kg)]] X 6.25,
24-Hour Urinary Protein
Changes in glomerular perm selectivity were assessed by comparing baseline with treatment values for the 24-hour urinary excretion of total protein, albumin, and IgG. Baseline total urinary protein excretion, which was the mean of two 24-hour urine collections measured 1 week apart prior to the start of the study, was 4.91 ± 0.96 g/
where [0.031 X weight (kg)] represents the nonurea nitrogen excretion. 33 Using this calculation, the range of protein intake was 42 to 125 g protein/d (0.57 to 1.67 g protein/kg/d), with a median of 66.3 g protein/d (0.88 g protein/ kg/d), and a mean of71 g protein/d (0.95 g protein/kg/d). Fig 1. (A) Twenty-fourhour urinary sodium excretion and (B) urinary urea nitrogen excretion were measured during the baseline period (-14) and at weekly intervals during both treatment periods. • , eEl; D, aI-antagonist. Values are mean ± SEM. To convert to mmol/d of urea, multiply by 35.70. No significant differences were detected.
250
12
200
10
~ N
ti-
E
150
100
A
T
i
-14
I
-7
I
0
i i i
7
STUDY D~Y
14
21
I
28
B
i i i
i i i
-14
-7
0
STUDY
14 D~Y
21
28
476
ROSENBERG AND HOSTETTER 110
110
100
100
*
90
T
I
BASELINE
I
7
I
14 STUDY DAY
I
I
21
28
90
B
T
I
BASELINE
24 h. No correlation was found between baseline levels of proteinuria and the response to the study drugs. A significant decrease in total urinary protein was seen with CEI on study days 7, 14, 21, and 28, compared with significant decrease with the lXI-antagonist only on study days 7 and 14 (Fig 3). At the end of the treatment periods, urinary protein excretion tended to be lower with the CEI compared with the lXI-antagonist (4.04 ± 1.00 v 4.39 ± 1.06 g/24 h; P = 0.06). Albumin excretion was significantly reduced from baseline values by both drugs, whereas a significant decrease in IgG excretion occurred only with the CEI (Fig 4). No significant difference was seen in the 24-hour collection on the final day for urinary albumin (CEI 2,338 ± 571 v lXI-antagonist 2,314 ± 543 mg/24 h; P = NS) or IgG (CEI 203 ± 60 v lXI-antagonist 216 ± 59 mg/24 h; P = NS). When the results of a subgroup of II patients, whose time-averaged MAP was not different between the study periods, were separately analyzed, a similar lack of difference between the drugs was seen. Changes in total urinary protein excretion were also analyzed by comparing the percent decrease in proteinuria from baseline to the absolute change in MAP (Fig 5). A total of 56 measurements were available for each parameter for both the CEI and lXI-antagonist treatment periods. As
I
7
I
I
I
14
21
28
STUDY DAY
Fig 2. (A) Supine MAP and (8) standing MAP were measured during the baseline period and at weekly intervals during the treatment periods. The baseline value is the average of four weekly readings. _, eEl; 0, iX1-antagonist. Values are mean ± SEM . •p < 0.05 v baseline. tP < 0.05 v lr1-antagonist.
shown in Fig 5, when a 21 to 30 mm Hg decrease in MAP occurred, the percent decrease in proteinuria was similar with the CEI and the lXI-antagonist, with the respective values being 31.6% ± 6.0% versus 28.6% ± 7.9%; P = NS. With lesser degrees of blood pressure reduction, a greater decrease in proteinuria occurred with the CEI. When an 11 to 20 mm Hg decrease in blood pressure occurred, the percent decrease in proteinuria was significantly greater on the CEI (33.1% ± 5.8% v 1.8% ± 7.8%; P < 0.01). When a 0 to 10 mm Hg decrease in blood pressure occurred, the respective values for the CEI and lXIantagonists were 30.9% ± 5.9% versus 9.3% ± 7.8%; P < 0.04. Although the numbers were small when either no change or an increase in blood pressure occurred, a trend favoring a greater reduction in proteinuria with CEI was found (7.2% ± 8.7% v - 4.4% ± 11.7%; P = NS). These results suggest that equivalent antiproteinuric effects of the CEI and lXI-antagonist only occur when blood pressure is markedly reduced, whereas for lesser degrees of blood pressure reduction the CEI had a more favorable effect on reducing proteinuria. To determine whether the changes in proteinuria could be due to changes in protein reabsorption, we examined the renal handling of fJr microglobulin. There was no difference in the
Table 4. Renal HemodynamiCS
Baseline
eEl IX1-antagonist
GFR (mL/s/m"j
RPF (mL/s/m"j
FF
0.52 ± 0.07 (31 ± 4) 0.40 ± 0.07* (24 ± 4) 0.43 ± 0.05* (26 ± 3)
3.22 ± 0.65 (193 ± 39) 2.85 ± 0.52 (171 ± 31) 2.62 ± 0 .45 (157 ± 27)
0.21 ± 0.02 0.18 ± 0.03 0.20 ± 0.02
NOTE. Values are mean ± SEM. Values in parentheses are in conventional units of mL/min/1 .73 m2 • P < 0.05 v baseline.
*
477
COMPARATIVE EFFECTS OF ANTIHYPERTENSIVES
6
20
(6)
-30 -40~--.-----~------~------.---
'"
~
~
2500
>-
-
""z
2000
A
""z
i
i
BASELINE
i
DAY 14
DAY 28
200
T
250
150
B
T
i
i
BASELINE
i
DAY 14
DAY 28
ROSENBERG AND HOSTETTER
478 7
6
5
M 0
I
..-
x '-' lSl
4 3
2
0
ALBUMIN
IgG
Fig 6. The fractional clearances of albumin and IgG were measured during the baseline period and on the final day of each treatment period. Values are mean ± SEM. *P < 0.05 v baseline. tP < 0.02 v a1-antagonist. D, baseline; ~, CEI; D, a1-antagonist.
tion was present between MAP and the fractional clearance of albumin or IgO for either drug.
Analysis of Patients Whose Baseline MAP Values Were in the Normotensive Range As seen in Table 1, the baseline MAP in seven patients was in the normotensive range (MAP < 107 mm Hg) and, thus, these patients would be considered to have adequate control of their hypertension. To determine whether a further reduction of blood pressure in these normotensive patients would be associated with any beneficial effects, we examined separately their response to CEI and ai-antagonist therapy (Table 5). CEI therapy significantly reduced 24-hour urinary protein, albumin, and IgO, as well as the fractional clearances of both albumin and IgO, whereas no significant change in these parameters was seen with ai-antagonist therapy. Also, a significant difference favoring the CEI was seen for the final day 24-hour urinary excretion of total protein and IgO, and the fractional clearances of albumin and IgO. DISCUSSION
Hypertension is a common, almost universal finding in the setting of chronic renal failure. I A faster rate of progression to end-stage renal disease occurs in those patients with higher blood pressures. 3 Therapy of hypertension can slow the rate of progression of renal disease. I ,16,19.27 The deleterious effects of poorly controlled hypertension are likely related to transmission of the
elevated systemic pressure to the glomerular microcirculation, as glomerular capillary hypertension is associated with more severe renal injury.4,5,34 Control of glomerular hypertension by dietary or pharmacological interventions lessens renal injury.4.5,7,8,34 Evidence derived from animal models suggest that certain antihypertensives are more efficacious than others. 5-8 For example, in the remnant kidney model, despite equivalent reductions in arterial pressure by either CEI or a combination of other agents, CEI reduced glomerular pressure more than the combination and resulted in less severe renal injury.5 A similar advantage of CEls over other agents has been demonstrated in experimental diabetes, and in spontaneously hypertensive rats and Dahl saltsensitive rats subjected to subtotal nephrectomy.6,8,9 In some studies, agents other than CEls afford protection only if control of systemic hypertension is associated with concomitant control of glomerular hypertension. 35 By contrast, recent studies by Yoshida et al have suggested that combinations of such standard agents can reduce histologic evidence of injury independent of changes in glomerular pressure, but the effects on proteinuria and long-term effects on OFR were not reported. 36 The ability ofCEls to reduce systemic and glomerular hypertension presumably depends on inhibition of angiotensin II formation. In the NS
100 80
60
40 20
MAP
0' ALBUMIN
(mm Hg)
(x 10- 4)
Fig 7. The MAP and fractional clearances of albumin and IgG are plotted after reanalysis of the data by excluding the three patients with the greatest differences in MAP between the two treatment periods. ~, CEI; D, a1-antagonist.
COMPARATIVE EFFECTS OF ANTIHYPERTENSIVES
479
Table 5. Patients With Baseline MAP in Normotensive Range
MAP(mm Hg) Urinary protein (g/24 h) Urinary albumin (g/24 h) Urinary IgG (mg/24 h) Albumin (X 10-3 ) IgG (X 10-3 ) GFR (mL/s/m2j
Baseline
GEl
I
N PATIENTS with chronic renal disease, control of hypertension reduces the rate of decay in renal function. I-3 Although a variety of antihypertensive agents have this beneficial effect, experimental evidence suggests that angiotensinconverting enzyme inhibitors (CEI) have a more specific effect on reducing renal injury, which may be related to their ability to lower glomerular capillary pressure. 4-9 Although numerous studies have examined the effects ofCEI in human renal disease,9-24 few studies have tested for a differential effect between CEI and other antihypertensives in humans. 17-24 The most definitive evidence for a beneficial effect of a particular antihypertensive on the progression of renal dis-
ease is dependent on demonstrating a decrease in the decline of glomerular filtration rate (GFR) and in progression to end stage. However, quantitating changes in permselectivity provides a useful shorter-term prognostic marker. 25-27 The purpose ofthe current study was to compare rigorously and directly the effects of two antihypertensives on renal function. Using a single-blinded, randomized, cross-over design, we compared the effects of an angiotensin-CEI (enalapril) and an lXI-antagonist (prazosin) on renal hemodynamics and glomerular permselectivity in patients with a variety of glomerular diseases. METHODS
From the Department ofMedicine, University ofMinnesota, Minneapolis, MN. Supported by National Institutes ofHealth Grants No. AM 31437 (TH.H.) and 5-M01-RR00400 (General Clinical Research Center), and by a grant from Merck Sharp and Dohme Research Laboratories. M.E.R. was supported by a research fellowship from the Medical Research Council of Canada. Address reprint requests to Mark E. Rosenberg, MD, University of Minnesota, Department of Medicine, Box 736 UMHC, 516 Delaware St SE, Minneapolis, MN 55455. © 1991 by the National Kidney Foundation, Inc. 0272-6386/91/1804-0007$3.00;0 472
Patients Fourteen patients were studied, nine with biopsy-proven primary glomerular disease, four with diabetic nephropathy as defined by proteinuric renal disease in the setting of longstanding diabetes mellitus and underlying retinopathy, and one who did not have a renal biopsy but clinically had primary renal disease (Table I). All patients were being treated for hypertension. Their antihypertensive medications, baseline mean arterial pressure (MAP), which was the average of four weekly readings, baseline creatinine clearances, and baseline proteinuria are shown in Table 1. Of note, seven patients had a MAP < 107 mm Hg, corresponding to a blood pressure of
American Journal of Kidney Diseases, Vol XVIII, No 4 (October), 1991: pp 472-482
COMPARATIVE EFFECTS OF ANTIHYPERTENSIVES
473
Table 1. Patient Characteristics
No.
Disease
Sex/Age
M/29
HTN
2 3 4
F/35 M/39 M/23
Diabetes Focal sclerosis Diabetes
5 6 7 8 9 10 11 12 13 14
M/29 F/32 F/46 F/39 M/65 M/62 M/46 F/30 F/31 M/34
Focal sclerosis Diabetes Membranous IgA nephropathy Membranous Diabetes Membranous SLE Focal sclerosis No biopsy
Antihypertensive Medications
Furosemide, metoprolol, minoxidil Furosemide Metoprolol Furosemide, metoprolol, hydralazine Bumetanide Furosemide Atenolol Furosemide, atenolol Furosemide Metolazone, clonidine, nifedipine Furosemide, verapamil Furosemide, propranolol Furosemide, verapamil Bumetanide
Baseline MAP (mmHg)
Baseline CrCI (mL/s/m~ [mL/min/ 1.73 m2l
Baseline Proteinuria (g/24h)
112
0.17 [18]
3.6
108 114 117
0.27 [28] 0.19 [20] 0.40 [42]
7.9 7.5 6.4
120 100 102 90 111 106 118 97 98 94
0.13 [14] 0.45 [47] 0.40 [42] 0.20 [21] 0.83 [86] 0.19 [20] 0.46 [48] 0.50 [52] 0.37 [38] 0.13[14]
5.2 0.9 1.2 0.7 8.5 13.1 3.6 2.5 4.6 2.4
Abbreviations: HTN, hypertension; SLE, systemic lupus erythematosus.
less than 140/90 and would, therefore, be considered to have adequate blood pressure control. 28 Entry criteria for the study were age between 18 and 65 years, serum creatinine between 133 and 442 JLmolfL/1.73 m 2 (1.5 and 5 mg/dL/1.73 m 2), dipstick-positive proteinuria, and no obvious reversible causes of their renal disease. If the patients were taking either of the two study medications, enalapril or prazosin, these were discontinued or substituted at least 4 weeks before the start of the study. Informed consent was obtained from all patients prior to entry into the study. The study was approved by the committee on human experimentation.
Design A single-blind, randomized, cross-over design was used. The total study duration was 15 weeks and consisted of a 3-week baseline period and two 4-week treatment periods, which were separated by a 4-week washout period. Throughout the study, the patients were seen at weekly intervals in the General Clinical Research Center (GCRC) at the University of Minnesota. The study began with a 3-week baseline period during which the patients were maintained on their previous antihypertensives (Table 1) and was followed by the first 4-week treatment period during which either the CEI (enalapriI) or the aI-antagonist (prazosin) was added to their previous antihypertensive medications. Following the first treatment period, the patients returned to their baseline medications for 4 weeks (washout period) before entering the second 4-week treatment period during which the second drug was added to their baseline medications. The order of the treatment periods was chosen by random assignment. Eight patients began with the CEI and six patients with the aI-antagonist. Both drugs were prepared in identical gelatin capsules insuring the single-blindedness of the study. The starting dose of the CEI was 2.5 mg
orally twice daily and of the aI-antagonist 2 mg orally twice daily. The goal of the treatment periods was to lower the MAP 10 mm Hg compared with baseline values. To insure equivalent reductions in blood pressure, the dosages of both drugs were adjusted on a weekly basis. On the night before the beginning and end of each treatment period, the patients were admitted to the GCRC. The following morning, blood was obtained in the supine position for measurement of plasma renin activity (PRA) and angiotensin-converting enzyme (ACE) activity. Measurement of PRA was repeated after I hour of standing. Renal clearance studies were then performed as outlined below. In all patients, weekly measurements of 24-hour urinary sodium, potassium, urea nitrogen, creatinine, and total protein; and serum sodium, potassium, chloride, bicarbonate, blood urea nitrogen (BUN), and creatinine were performed during all baseline and treatment visits. Hemoglobin, white blood cell (WBC) count, and 24-hour urinary albumin and IgG were measured every other week. Baseline values are those obtained before addition of the study drugs.
Blood Pressure Measurement Blood pressure was measured in an identical fashion throughout the study by a group of nurses who were blinded as to the study drug. Two blood pressure measurements 5 minutes apart were made in the supine position after 10 minutes of rest. Two standing blood pressure measurements 5 minutes apart were than performed. MAP was calculated by adding one third of the pulse pressure to the diastolic pressure. The mean value for MAP of all four readings was used to adjust the dose of the study drugs.
Clearance Studies Clearance studies were performed between 8:00 AM and noon at the beginning and end of each treatment period as
474
ROSENBERG AND HOSTETTER
Table 2. General Parameters
Weight (kg) Sodium (mmoI/L) Potassium (mmol/L) Urea nitrogen (mmol/L) [mg/dL] Creatinine (ILmoI/L) [mg/dL] CrCI (mL/s/m 2 ) [mL/min/1.73 m2] Total protein (gIL) [mg/dL] Albumin (gIL) [mg/dL] IgG (gIL [mg/dL] Hemoglobin (gIL) [g/dL]
Baseline
GEl
a,-Antagonist
75 ± 4 139 ± 1 4.2 ± 0 .2 16.0 ± 1.4 [45 ± 4] 283 ± 44 [3.2 ± 0.5] 0.35 ± 0.05 [36 ± 5] 58 ± 2 [5.8 ± 0 2] . 31 ± 2 [3.1 ± 0.2) 8.6 ± 0.8 [858 ± 76] 119±3 [11.9±0.3]
74 ± 4 138 ± 1 4.6 ± 0.1 18.6 ± 2.1 [52 ± 6] 301 ± 44 [3.4 ± 0.5] 0.33 ± 0.06 [34 ± 6] 59 ± 2 [5.9 ± 0.2] 30 ± 2 [3.0 ± 0.2] 8.6 ± 0.8 [856 ± 78] 116 ± 4 [11.6 ± 0.4]
76 ± 4 139 ± 1 4.1 ± 0.1 16.4±1 .4 [46±4] 292 ± 44 [3.3 ± 0.5] 0.37 ± 0.06 [38 ± 6] 58 ± 2 [5.8 ± 0.2] 31 ± 2 [3.1 ± 0.20) 8.5 ± 0.6 [853 ± 64] 117±4 [11 .7 ± 0.4]
NOTE. Values are mean ± SEM. No significant differences were detected.
described previously.29 Two intravenous catheters were inserted, one for the infusion of inulin (lso-Tex Diagnostics, Friendswood, TX) and p-aminohippurate (PAH) (Merck Sharpe & Dohme, West Point, PAl and the other in the opposite arm for repeat blood sampling. Throughout the clearance studies, the patients drank 250 mL of tap water every 30 minutes to maintain adequate urine flow. A loading dose of inulin and PAH was infused over 10 minutes followed by a continuous infusion of inulin and PAH designed to maintain blood levels of25 mgjdL and 2 mgjdL, respectively. Following a 60-minute equilibration period, three 30-minute urine collections were performed. Blood pressure and a blood sample for measurement of inulin and PAH were obtained at the midpoint of each clearance period. During the second clearance period, blood and urine were also obtained for measurement of albumin and IgG to enable calculation of the fractional clearances of these proteins. The fractional clearances were calculated by dividing the clearance of the protein by the clearance of the filtration marker inulin. The reabsorption of tl2-microglobulin was calculated by subtracting the urinary excretion of tl2-microglobulin from its filtered load. The average value for inulin and PAH clearances during the three periods was considered to represent GFR and renal plasma flow (RPF), respectively. PAH clearance was corrected for published values of extraction ratio for patients with renal disease by dividing the PAH clearance by 0.7. 30
Laboratory Measurements Urine and heparinized plasma were used for colorimetric measurements of inulin and P AH. 31 .32 Albumin and IgG were measured by rate nephelometry (Beckman Immunochemistry System; Beckman Instruments, Brea, CAl. Total protein was measured by the biuret technique. An enzyme immunoassay (Phadezym tlrmicroTest; Pharmacia Diagnostics, Piscataway, NJ) was used to determine tlrmicroglobulin. Serum sodium, potassium, chloride, bicarbonate, glucose, calcium, urea nitrogen, creatinine, and urine sodium, potassium, urea nitrogen, and creatinine were measured by an autoanalyzer (Beckman ASTRA). Hemoglobin and WBC count were determined by coulter counter. PRA was determined by radioimmunoassay for angiotensin I at pH 5.7 using a commercially available kit (DuPont Diagnostic Imaging, North Billerica, MA). Serum ACE activity
was measured using the ACE Microvial Radioassay system (Ventrex Laboratories, Portland, ME).
Statistics All data are reported as the mean ± SEM. Statistical differences between baseline and treatment values and between CEI and ai-antagonist values were calculated by either the Student's paired t test or the Wilcoxon signed rank test depending on whether the values were normally distributed or not. A statistically significant difference between values was considered for a P value less than 0.05.
RESULTS
General Parameters
The average daily dose (mg/d) of the CEI was week 1,5.0 ± 0.0; week 2,5.7 ± 0.6; week 3, 6.6 ± 0.9; week 4, 7.0 ± l.0; and of the aI-antagonist, week 1,3.9 ± 0.1; week 2,5.1 ± 0.5; week 3, 6.6 ± 1.0; week 4, 7.6 ± 104. Hyperkalemia requiring potassium-binder therapy occurred transiently in one patient on the CEI and two patients on the ai-antagonist. Other side effects included dyspepsia in one patient while taking the CEI and orthostatic hypotension, and nasal stuffiness each occurring in one patient on aI-antagonist therapy. No significant differences in serum sodium, potassium, BUN, creatinine, total protein, albumin, IgG, hemoglobin, creatinine clearance, or weight were seen throughout the study, although serum potassium tended to be higher with the CEI (Table 2). Serum chloride, bicarbonate, blood sugar, and WBC count were also similar throughout the study (data not shown). ACE activity was significantly lower and supine and standing PRA significantly higher at the end of the CEI period compared with both baseline and aI-antagonist values, establishing the patients'
475
COMPARATIVE EFFECTS OF ANTIHYPERTENSIVES Table 3. Angiotensin-Converting Enzyme and Plasma Renin Activity
Hemodynamics
Consistent with the goals of the study, both supine and standing MAP were significantly reduced by both drugs compared with baseline MAP except for the supine day 14 MAP of the ai-antagonist, when a numerical but not a statistical difference was seen (Fig 2). At the end of the treatment periods, there was no difference between the two drugs for either supine MAP (CEI 95 ± 4 val-antagonist 99 ± 3 mm Hg; P = NS) or standing MAP (CEI 94 ± 4 val-antagonist 95 ± 4 mm Hg; P = NS). As an estimate of the time-averaged MAP throughout the study, we added the individual MAP readings from day 7, 14,21, and 27 and divided by 4. This estimate of time-averaged MAP was lower with the CEI compared with the ai-antagonist (94 ± 3 v 98 ± 3 mm Hg; P < 0.05). As well, during the clearance period MAP was lower with CEI compared with ai-antagonist (CEI 101 ± 4 val-antagonist 108 ± 4 mm Hg; P < 0.05) despite administration of both drugs on the morning of the study and identical clearance protocols. No significant differences were seen for GFR, RPF, or filtration fraction between the CEI and ai-antagonist. Compared with baseline values, a significant decrease in GFR occurred with both drugs (Table 4). No correlation was present between changes in GFR and the decrease in blood pressure.
PRA (ng AI/mL/h) ACE (U/mL)
Supine
Standing
Baseline 4.51 ± 0.41 1.92 ± 0.33 5.41 ± 1.97 CEI 0.20 ± 0.02*t 9.05 ± 3.08*t 12.71 ± 4.00*t al-antagonist4.41 ± 0.37 1.79 ± 0.34 4.13 ± 1.14 NOTE. Values are mean ± SEM. P < 0.05 v baseline. t P < 0.05 val-antagonist. *
compliance with CEI (Table 3). Urinary excretion of sodium and urinary urea nitrogen were similar throughout the study, indicating constancy of dietary sodium and protein intake (Fig 1). Using all of the values for 24-hour urinary sodium measured throughout the study, the range of sodium excretion was 48 to 435 mmol/d, with a median of 154 mmol/d, and a mean ± SE of 166 ± 5 mmol/d. Similarly, the range of 24-hour urinary urea excretion ranged between 160 to 630 mmol/d (4.48 to 17.65 g nitrogen [N]/d), with a median of296 mmol/d (8.29 g N/d), and a mean ± SE of 323 ± 9 mmol/d (9.04 ± 0.24 g N/d). Assuming the patients were in steady-state protein balance, estimated protein intakes were calculated based on the following formula: Protein Intake
=
[UUN (g N/24 h)
X [0.031 X weight (kg)]] X 6.25,
24-Hour Urinary Protein
Changes in glomerular perm selectivity were assessed by comparing baseline with treatment values for the 24-hour urinary excretion of total protein, albumin, and IgG. Baseline total urinary protein excretion, which was the mean of two 24-hour urine collections measured 1 week apart prior to the start of the study, was 4.91 ± 0.96 g/
where [0.031 X weight (kg)] represents the nonurea nitrogen excretion. 33 Using this calculation, the range of protein intake was 42 to 125 g protein/d (0.57 to 1.67 g protein/kg/d), with a median of 66.3 g protein/d (0.88 g protein/ kg/d), and a mean of71 g protein/d (0.95 g protein/kg/d). Fig 1. (A) Twenty-fourhour urinary sodium excretion and (B) urinary urea nitrogen excretion were measured during the baseline period (-14) and at weekly intervals during both treatment periods. • , eEl; D, aI-antagonist. Values are mean ± SEM. To convert to mmol/d of urea, multiply by 35.70. No significant differences were detected.
250
12
200
10
~ N
ti-
E
150
100
A
T
i
-14
I
-7
I
0
i i i
7
STUDY D~Y
14
21
I
28
B
i i i
i i i
-14
-7
0
STUDY
14 D~Y
21
28
476
ROSENBERG AND HOSTETTER 110
110
100
100
*
90
T
I
BASELINE
I
7
I
14 STUDY DAY
I
I
21
28
90
B
T
I
BASELINE
24 h. No correlation was found between baseline levels of proteinuria and the response to the study drugs. A significant decrease in total urinary protein was seen with CEI on study days 7, 14, 21, and 28, compared with significant decrease with the lXI-antagonist only on study days 7 and 14 (Fig 3). At the end of the treatment periods, urinary protein excretion tended to be lower with the CEI compared with the lXI-antagonist (4.04 ± 1.00 v 4.39 ± 1.06 g/24 h; P = 0.06). Albumin excretion was significantly reduced from baseline values by both drugs, whereas a significant decrease in IgG excretion occurred only with the CEI (Fig 4). No significant difference was seen in the 24-hour collection on the final day for urinary albumin (CEI 2,338 ± 571 v lXI-antagonist 2,314 ± 543 mg/24 h; P = NS) or IgG (CEI 203 ± 60 v lXI-antagonist 216 ± 59 mg/24 h; P = NS). When the results of a subgroup of II patients, whose time-averaged MAP was not different between the study periods, were separately analyzed, a similar lack of difference between the drugs was seen. Changes in total urinary protein excretion were also analyzed by comparing the percent decrease in proteinuria from baseline to the absolute change in MAP (Fig 5). A total of 56 measurements were available for each parameter for both the CEI and lXI-antagonist treatment periods. As
I
7
I
I
I
14
21
28
STUDY DAY
Fig 2. (A) Supine MAP and (8) standing MAP were measured during the baseline period and at weekly intervals during the treatment periods. The baseline value is the average of four weekly readings. _, eEl; 0, iX1-antagonist. Values are mean ± SEM . •p < 0.05 v baseline. tP < 0.05 v lr1-antagonist.
shown in Fig 5, when a 21 to 30 mm Hg decrease in MAP occurred, the percent decrease in proteinuria was similar with the CEI and the lXI-antagonist, with the respective values being 31.6% ± 6.0% versus 28.6% ± 7.9%; P = NS. With lesser degrees of blood pressure reduction, a greater decrease in proteinuria occurred with the CEI. When an 11 to 20 mm Hg decrease in blood pressure occurred, the percent decrease in proteinuria was significantly greater on the CEI (33.1% ± 5.8% v 1.8% ± 7.8%; P < 0.01). When a 0 to 10 mm Hg decrease in blood pressure occurred, the respective values for the CEI and lXIantagonists were 30.9% ± 5.9% versus 9.3% ± 7.8%; P < 0.04. Although the numbers were small when either no change or an increase in blood pressure occurred, a trend favoring a greater reduction in proteinuria with CEI was found (7.2% ± 8.7% v - 4.4% ± 11.7%; P = NS). These results suggest that equivalent antiproteinuric effects of the CEI and lXI-antagonist only occur when blood pressure is markedly reduced, whereas for lesser degrees of blood pressure reduction the CEI had a more favorable effect on reducing proteinuria. To determine whether the changes in proteinuria could be due to changes in protein reabsorption, we examined the renal handling of fJr microglobulin. There was no difference in the
Table 4. Renal HemodynamiCS
Baseline
eEl IX1-antagonist
GFR (mL/s/m"j
RPF (mL/s/m"j
FF
0.52 ± 0.07 (31 ± 4) 0.40 ± 0.07* (24 ± 4) 0.43 ± 0.05* (26 ± 3)
3.22 ± 0.65 (193 ± 39) 2.85 ± 0.52 (171 ± 31) 2.62 ± 0 .45 (157 ± 27)
0.21 ± 0.02 0.18 ± 0.03 0.20 ± 0.02
NOTE. Values are mean ± SEM. Values in parentheses are in conventional units of mL/min/1 .73 m2 • P < 0.05 v baseline.
*
477
COMPARATIVE EFFECTS OF ANTIHYPERTENSIVES
6
20
(6)
-30 -40~--.-----~------~------.---
'"
~
~
2500
>-
-
""z
2000
A
""z
i
i
BASELINE
i
DAY 14
DAY 28
200
T
250
150
B
T
i
i
BASELINE
i
DAY 14
DAY 28
ROSENBERG AND HOSTETTER
478 7
6
5
M 0
I
..-
x '-' lSl
4 3
2
0
ALBUMIN
IgG
Fig 6. The fractional clearances of albumin and IgG were measured during the baseline period and on the final day of each treatment period. Values are mean ± SEM. *P < 0.05 v baseline. tP < 0.02 v a1-antagonist. D, baseline; ~, CEI; D, a1-antagonist.
tion was present between MAP and the fractional clearance of albumin or IgO for either drug.
Analysis of Patients Whose Baseline MAP Values Were in the Normotensive Range As seen in Table 1, the baseline MAP in seven patients was in the normotensive range (MAP < 107 mm Hg) and, thus, these patients would be considered to have adequate control of their hypertension. To determine whether a further reduction of blood pressure in these normotensive patients would be associated with any beneficial effects, we examined separately their response to CEI and ai-antagonist therapy (Table 5). CEI therapy significantly reduced 24-hour urinary protein, albumin, and IgO, as well as the fractional clearances of both albumin and IgO, whereas no significant change in these parameters was seen with ai-antagonist therapy. Also, a significant difference favoring the CEI was seen for the final day 24-hour urinary excretion of total protein and IgO, and the fractional clearances of albumin and IgO. DISCUSSION
Hypertension is a common, almost universal finding in the setting of chronic renal failure. I A faster rate of progression to end-stage renal disease occurs in those patients with higher blood pressures. 3 Therapy of hypertension can slow the rate of progression of renal disease. I ,16,19.27 The deleterious effects of poorly controlled hypertension are likely related to transmission of the
elevated systemic pressure to the glomerular microcirculation, as glomerular capillary hypertension is associated with more severe renal injury.4,5,34 Control of glomerular hypertension by dietary or pharmacological interventions lessens renal injury.4.5,7,8,34 Evidence derived from animal models suggest that certain antihypertensives are more efficacious than others. 5-8 For example, in the remnant kidney model, despite equivalent reductions in arterial pressure by either CEI or a combination of other agents, CEI reduced glomerular pressure more than the combination and resulted in less severe renal injury.5 A similar advantage of CEls over other agents has been demonstrated in experimental diabetes, and in spontaneously hypertensive rats and Dahl saltsensitive rats subjected to subtotal nephrectomy.6,8,9 In some studies, agents other than CEls afford protection only if control of systemic hypertension is associated with concomitant control of glomerular hypertension. 35 By contrast, recent studies by Yoshida et al have suggested that combinations of such standard agents can reduce histologic evidence of injury independent of changes in glomerular pressure, but the effects on proteinuria and long-term effects on OFR were not reported. 36 The ability ofCEls to reduce systemic and glomerular hypertension presumably depends on inhibition of angiotensin II formation. In the NS
100 80
60
40 20
MAP
0' ALBUMIN
(mm Hg)
(x 10- 4)
Fig 7. The MAP and fractional clearances of albumin and IgG are plotted after reanalysis of the data by excluding the three patients with the greatest differences in MAP between the two treatment periods. ~, CEI; D, a1-antagonist.
COMPARATIVE EFFECTS OF ANTIHYPERTENSIVES
479
Table 5. Patients With Baseline MAP in Normotensive Range
MAP(mm Hg) Urinary protein (g/24 h) Urinary albumin (g/24 h) Urinary IgG (mg/24 h) Albumin (X 10-3 ) IgG (X 10-3 ) GFR (mL/s/m2j
Baseline
GEl
