Peripheral and renal vein renin activity in patients with renovascular hypertension due to nonspecific aortoarteritis.

The diagnostic utility of peripheral and renal vein renin estimations in relation to angiographic findings was evaluated in 13 patients with renovascular hypertension and non-specific aortoarteritis (NSAA, Gr I), in comparison with 10 patients with renal artery stenosis due to other causes (Gr II). Plasma renin activity (PRA) was measured by radioimmunoassay. Blood samples were collected after angiography from the femoral vein and renal vein on the affected side followed by sampling from the less affected or unaffected side. Renal vein renin ratio (RVRR) was calculated from renal vein renin values. The effect of captopril (25 mg oral) on blood pressure, PRA, and RVRR was examined in 8 patients from each group. Normotensive volunteers (8) with moderately low salt intake were also included in the study for comparison of twenty-four-hour urinary sodium output, peripheral PRA, and response to captopril. The mean peripheral PRA was high in both groups as compared with normotensive controls; however, the values were lower in patients with NSAA. The rise in PRA in response to captopril was insignificant in Gr I (p greater than 0.05) and RVRR greater than 1.5 was observed in 5 of 13 patients in contrast to 9 of 10 in Gr II (p less than 0.05). A paradoxical ratio, ie, (high renal vein renin levels on the less stenotic side) was noticed in 3 patients of Gr I, whereas none of the patients of GR II showed such a ratio. An improvement in RVRR after captopril was observed in 50% of patients of Gr I as compared with a marked response in all patients of Gr II.(ABSTRACT TRUNCATED AT 250 WORDS)     

Effects of converting enzyme inhibition on split renal function in renovascular hypertension

The effects of captopril on effective renal plasma flow and glomerular filtration rate were studied using a noninvasive radioisotopic method on individual kidneys in eight patients with renovascular hypertension and 12 patients with essential hypertension with various renin levels. Four patients with renovascular hypertension had unilateral while three had bilateral renal artery stenosis. The effective renal plasma flow and glomerular filtration rate were determined by using 131I-iodohippurate sodium and 99mTc-diethylenetriamine pentaacetic acid, respectively. Glomerular filtration rate and effective renal plasma flow were significantly reduced in the stenotic kidneys of patients with renovascular hypertension compared with values in nonstenotic kidneys (p less than 0.01). Treatment with captopril, 37.5 to 75 mg/day for 1 to 48 weeks, further reduced the glomerular filtration rate only in stenotic kidneys, and effective renal plasma flow increased in both kidney types. In two of the three renal hypertensive patients with bilateral renal artery stenosis, captopril produced a reversible azotemia that was unrelated to the fall in blood pressure, as evidenced by the lack of azotemia seen after a moderate blood pressure reduction induced by other antihypertensive medications. These results indicate that endogenous angiotensin II is essential in maintaining the glomerular filtration rate in stenotic kidneys and suggest that a reduction in glomerular filtration rate during captopril administration could indicate the presence of renal artery stenosis.     

Predictive value and changes of renin secretion in hypertensive patients with unilateral renovascular disease undergoing successful renal angioplasty

Peripheral and renal vein renin activity was measured in 46 hypertensive patients with arteriographically diagnosed unilateral renal artery stenosis before and six months after technically successful renal angioplasty. The renin-sodium profile was high in 80 percent before angioplasty, fell in all patients, and became normal or low in 85 percent after angioplasty. Renal vein renin activity, which initially showed lateralization of renin secretion to the ischemic kidney with contralateral suppression, became normal. Renal vein renin activity was more reliable for predicting the therapeutic response when expressed as the increment of renin for each renal vein over the infrarenal vena caval value (sensitivity 74 percent, specificity 100 percent) than as the ratio between the two renal veins (sensitivity 62 percent, specificity 60 percent). The predictive value of renal vein renin activity is poor when plasma renin activity is stimulated by long-term administration of captopril. These data support the usefulness and define the limitations of peripheral and renal vein renin measurements in selecting patients for treatment by renal angioplasty.     

Renal venous renin in patients with renovascular hypertension.

Renal venous and peripheral plasma renin activities were determined in 29 operated patients with renovascular hypertension and in 10 patients with essential hypertension. The majority of patients with renovascular hypertension exhibited elevated peripheral plasma renin activity, but the most striking increase of renin activity was demonstrated in the venous effluent of the involved kidney. Using data obtained in patients with essential hypertension, the ratio of renal vein renin activity not exceeding 1.4 was assumed normal. In patients with renovascular hypertension, the values above 1.4 were accepted as lateralizing ratios. In 78.6 % of patients with unilateral renal artery stenosis and a lateralizing renal vein renin ratio, normotension or improvement of blood pressure control were obtained post-operatively. The discussion emphasis the importances of renal vein renin estimations with the calculation of renal vein ratio for determining the functional significance of renal artery stenosis and for predicting the surgical outcome     

Inverse relation of exchangeable sodium and blood pressure in hypertensive patients with renal artery stenosis

Measurements of exchangeable sodium, arterial pressure and plasma concentrations of active renin, angiotensin II, aldosterone, sodium and potassium were made in 35 hypertensive patients with renal artery stenosis, 30 having unilateral renal arterial lesions. Plasma urea was below 7 mmol/l in 24 of the patients with unilateral lesions. In these and in the whole group of 35 patients there were significant inverse correlations between exchangeable sodium and diastolic blood pressure and between plasma sodium concentration and diastolic pressure. Six patients had hyponatraemia with a plasma sodium concentration less than 135 mmol/l. All were sodium-deplete with secondary hyperaldosteronism, three also having malignant-phase hypertension. Twelve of the patients with unilateral renal artery stenosis underwent bilateral ureteric catheterization. Sodium excretion was greater from the contralateral kidney than from the affected kidney and the rate of sodium excretion from the former, but not from the latter, was significantly related to arterial pressure. The relation of diastolic blood pressure and exchangeable sodium is the opposite of the positive correlation found in essential hypertension and Conn's syndrome. In renal artery stenosis the inverse correlation could result from a natriuretic effect of increased arterial pressure occurring mainly in the contralateral kidney.     

Captopril in the management of hypertension with renal artery stenosis: Its long-term effect as a predictor of surgical outcome

Fifteen patients with hypertension and unilateral renal artery disease were treated with captopril alone; 10 came to operation and were later assessed postoperatively with no drug treatment. Captopril caused both immediate and sustained decreases in plasma angiotensin II and aldosterone, with increases in plasma active renin and blood angiotensin I concentrations. Decrements in systolic and diastolic pressure 2 hours after the first dose of captopril were closely correlated with the initial decreases in plasma angiotensin II. Blood pressure was decreased by long-term captopril therapy irrespective of whether plasma angiotensin II was abnormally high before treatment. The long-term response of both systolic and diastolic pressure correlated well with the response to surgery. By contrast, the blood pressure decrease 2 hours after the initial dose of captopril variously underestimated and overestimated the decrease during prolonged use of the drug and did not relate to surgical outcome. In patients who, before treatment, had secondary aldosteronism, hyponatremia, hypokalemia and sodium and potassium deficiency, captopril corrected these abnormalities. In the remaining patients, long-term captopril therapy did not alter exchangeable sodium, plasma sodium or total body potassium, although plasma potassium levels increased.     

Different effects of captopril on blood pressure in the acute and chronic two-kidney Goldblatt hypertensive dogs

Captopril was administered to acute (8 to 14 days after unilateral renal artery constriction) and chronic (71 to 127 days after the constriction) two-kidney Goldblatt hypertensive dogs, and to normotensive ones for 21 days (oral administration of 10, 20 and 40 mg/kg/day, consecutively each 7-day period). The decrease of arterial blood pressure was remarkable in hypertensive animals with high plasma renin activity, but not in the normotensive animals. In the acute stage of hypertension, the antihypertensive effect of captopril was dose-dependent and persistent even after its cessation. In the chronic stage of hypertension, blood pressure also decreased, but the response was not dose-dependent and did not continue after cessation. Plasma renin activity rose in both hypertensive and normotensive animals during the treatment with captopril. There were no significant changes in heart rate, daily urinary volume, sodium balance, and renal clearances of sodium (CNa), potassium (CK), chloride (CCl) and creatinine (CCr). Circulating blood volume was also not altered. These results indicate that the main mechanism of antihypertensive effect of captopril in two-kidney Goldblatt hypertensive dogs is an inhibition of the angiotensin converting enzyme. In addition, the different effects in the acute and chronic hypertensive dogs suggest that some differences exist in the mechanism(s) of maintaining blood pressure between the two stages of two-kidney Goldblatt hypertension in dogs.     

Aspirin lowers blood pressure in patients with renovascular hypertension

To clarify the role of renal prostanoid in hyperreninemia and high blood pressure in human renovascular hypertension, we measured prostaglandin E2 and renin activity in renal venous and abdominal aortic plasma before and after the intravenous administration of the cyclooxygenase inhibitor, aspirin DL-lysine. Subjects were six patients with unilateral renovascular hypertension and six with essential hypertension. In patients with renovascular hypertension, prostaglandin E2 concentration in renal venous plasma from the stenotic kidney was 9.25 +/- 1.48 pg/ml, which was significantly higher (p less than 0.01) than the concentration in the renal venous plasma from the normal kidney (4.97 +/- 1.02 pg/ml) or in the aortic plasma (2.59 +/- 0.15 pg/ml). Plasma renin activity was also higher in the renal vein of the stenotic kidney than in the other two sites. The stenotic side/normal side ratio of the renal venous prostaglandin E2 correlated significantly with a renin ratio greater than 1.5 (r = 0.8211, p less than 0.05). Intravenous injection of aspirin DL-lysine (18 mg/kg) 30 minutes later markedly suppressed prostaglandin E2 and renin levels at all sites and clearly lowered arterial blood pressure (mean: from 120 +/- 6 to 110 +/- 5 mm Hg, p less than 0.01). The reduction in blood pressure correlated significantly with the suppression of plasma renin activity in the aorta (p less than 0.05) and in the renal vein of the stenotic kidney (p less than 0.01). Conversely, in patients with essential hypertension, aspirin had little effect on renin levels and increased mean blood pressure.(ABSTRACT TRUNCATED AT 250 WORDS)     

Hypertension with renal arterial stenosis: humoral, hemodynamic and histopathologic factors.

In 46 hypertensive patients with unilateral renal arterial stenosis, peripheral and renal venous plasma renin activity, juxtaglomerular cell count and granularity and systolic pressure gradient across the stenosis were determined. After corrective surgery and a mean postoperative observation period of 4.3 years, 18 patients were completely relieved of hypertension (good responders), 14 had a substantial reduction in arterial pressure (fair responders) and 14 remained hypertensive (poor responders). Analysis of plasma renin activity in both renal veins indicated that a ratio (stenotic/nonstenotic side) greater than 2.0 correctly predicted a favorable surgical result in all cases. Peripheral plasma renin activity was greater than normal in 65 percent of good responders, in 50 percent of fair responders and in one nonresponder. The prognostic accuracy of a pressure gradient greater than 40 mm Hg was 78 percent; no patient with a gradient of less than 40 mm Hg benefited from surgery. An increased juxtaglomerular cell count on the affected side predicted a successful operative result in 88 percent, as did increased granularity in 85 percent of cases. Renal venous renin ratio correlated positively (r =0.738, P less than 0.001) with the pressure gradient across the stenosis. The renal venous plasma renin activity of the affected side also correlated positively (r = 0.771, P less than 0.001) with the absolute count of granular cells in the juxtaglomerular apparatus. Plasma renin activity in both renal veins is the most reliable predictor of operative outcome. The addition of juxtaglomerular cell count or pressure gradient across the stenosis increases prognostic accuracy only slightly. The close mutual correlations between renal venous renin ratio, pressure gradient and juxtaglomerular cell count support the experimental evidence of a causal relation between the hemodynamic effects of the arterial lesion and the humoral and histologic changes observed in hypertension with renal arterial stenosis.     

Mechanism of the blood pressure and renal hemodynamic effects of captopril

This study was designed to investigate the mechanisms of captopril's chronic effect on arterial pressure and renal function. In dogs maintained on high sodium intake (250 mEq/day), 6 days of captopril infusion caused no change in arterial pressure, renal hemodynamics, sodium excretion or plasma aldosterone concentration. Infusion of captopril for 7 days also caused no significant changes in arterial pressure or renal function in dogs made hypertensive by chronic infusion of angiotensin II and high sodium intake, a model of hypertension in which plasma renin activity is undetectable and prostaglandin and bradykinin formation may be elevated. In dogs maintained on low sodium intake, chronic infusion of captopril decreased arterial pressure and plasma aldosterone concentration markedly while increasing effective renal plasma flow. Infusion of aldosterone (200 μg/day) for 8 days during captopril infusion restored plasma aldosterone concentration but did not significantly change arterial pressure or renal function, indicating that decreased plasma aldosterone concentration did not play a major role in the hypotensive and renal effects of captopril. However, angiotensin II infusion (10 ng/kg/min) for 8 days during captopril infusion restored arterial pressure, plasma aldosterone concentration and renal function toward control levels. These data suggest that the effects of captopril on arterial pressure, renal hemodynamics and electrolyte excretion are mediated primarily by decreased angiotensin II formation.     

Captopril in clinical hypertension. Changes in components of renin-angiotensin system and in body composition in relation to fall in blood pressure with a note on measurement of angiotensin II during converting enzyme inhibition.

The effect of the converting enzyme inhibitor captopril on arterial pressure, the components of the renin-angiotensin-aldosterone system, and body sodium and potassium content was studied in eight hypertensive patients with renal artery stenosis and, in conjunction with diuretics, in seven patients with hypertension unresponsive to previous treatment. Two hours after the first dose, captopril caused significant falls in systolic and diastolic pressures, plasma angiotensin II, and aldosterone, with converse increases in angiotensin I and both active and total renin; the initial fall in diastolic pressure was significantly related to the drop in plasma angiotensin II. The biochemical changes were sustained during prolonged treatment, even when diuretics were added. One untreated patient with renal artery occlusion had severe secondary aldosterone excess, was sodium and potassium depleted, and severely hyponatraemic and hypokalaemic; captopril restored blood pressure, plasma electrolyte concentrations, and exchangeable sodium and total body potassium to normal. In one man with renal artery stenosis and overall renal impairment captopril led to sodium retention, and blood pressure did not fall until a diuretic was added. In the remaining patients with renal artery stenosis, pretreatment renin, angio tensin II, and aldosterone concentrations were either normal or only modestly raised, and plasma electrolyte concentrations and body content of sodium and potassium were normal. Captopril alone controlled arterial pressure in all, three cases showing a gradual fall of pressure over the first six weeks of treatment; no significant changes in exchangeable sodium or total body potassium were seen. The group of patients with previously intractable hypertension were all controlled with a combination of captopril and diuretic.     

Active and inactive renin after a single dose of captopril in hypertensive patients

In 25 hypertensive patients (15 with renal artery stenosis and 10 with essential hypertension), captopril, in a single 12.5 mg dose, caused a prompt decrease in arterial pressure without changing the heart rate. Plasma active and trypsin-activated renin significantly increased, whereas inactive renin and plasma aldosterone decreased. The plasma active/ inactive renin ratio was also increased, suggesting that captopril, together with a release of active renin, may induce an in vivo activation of inactive renin.No correlations were found between blood pressure changes and both pretreatment and captopril-induced variations of active, inactive and trypsin-activated renin or the active/inactive ratio. However, the percent decrease in mean arterial pressure was significantly related to the increase in the active/inactive renin ratio in a group of patients whose blood pressure was brought to normal (r = ?0.78; p < 0.001). This finding suggests the possibility that vasodilating substances, in addition to inhibiting angiotensin II formation, might play some role both in exerting a full effect of captopril on blood pressure and in triggering the in vivo mechanisms of inactive renin activation.     

Is single oral administration of captopril beneficial in screening for primary aldosteronism?

Plasma renin activity (PRA) and plasma aldosterone concentration (PAC) were measured in 19 cases with primary aldosteronism (PA) and in 72 with essential hypertension (EHT) to differentiate the two disorders during the following conditions on normal salt diet: after overnight recumbency (basal state) and 2 hours after oral administration of 25 mg of captopril. Screening criteria were determined so that all PA patients were diagnosed as positive, and their specificities were compared with those of other conventional screening methods for PA. After captopril administration, the specificity of a criterion based on a combination of PAC and PAC/PRA ratio was 93% and positive predictive value was 79%. This criterion was superior to blood pressure response to angiotension II analog infusion, PRA on salt depletion, and to PAC on salt loading. However, higher specificity (97%) and positive predictive value (90%) were obtained from a criterion based on a combination of basal PAC and PAC/PRA ratio. Therefore, the use of a combination criterion based on PAC and PAC/PRA ratio at basal state rather than after captopril administration may give a satisfactory result in the screening for PA.     

The Effects of Saralasin, an Angiotensin II Antagonist on Blood Pressure and the Renin-Angiotensin-Aldosterone System in Normal and Hypertensive Subjects

Summary: The effects of saralasin, an angiotensin II antagonist, on blood pressure and the renin-angiotensin-aldosterone system in recumbent normal and hypertensive subjects.
Blood pressure reduction with saralasin infusion was seen only in hypertensive patients with abnormally elevated basal plasma renin and angiotensin II levels, and after sodium depletion the reduction in blood pressure was more marked. In normal subjects, and in hypertensives with plasma renin and angiotensin II levels within the normal range, there was no marked fall in blood pressure across saralasin infusion regardless of the sodium status of the individual.
Plasma aldosterone concentration fell during saralasin infusion in those subjects with high baseline renin and angiotensin II levels. This fall occurred in the sodium replete and deplete states. In the normal subjects, and those hypertensives with normal plasma renin levels, there was no fall in aldosterone in the sodium replete state. However, after sodium depletion the expected rise in aldosterone was abolished during saralasin infusion, the plasma aldosterone falling to within the normal sodium replete range, rising again after the saralasin infusion was stopped.
This study supports the concept of a direct role for renin and angiotensin II in the maintenance of hypertension in those subjects with elevated basal plasma renin. Plasma aldosterone would appear to be controlled, at least in part, by the prevailing plasma angiotensin II level in those subjects with elevated basal levels of angiotensin II; that is in high renin hypertensives, and in normal subjects and normal renin hypertensives who are sodium deplete.     

The captopril test for identifying renovascular disease in hypertensive patients

To develop a screening test for identifying renovascular hypertension, the blood pressure and plasma renin activity responses to an oral test dose of captopril were studied in 246 quietly seated hypertensive patients. The following criteria were developed that exploit the hyperresponsiveness of renin secretion in renovascular hypertensive patients: a 60-minute post-captopril plasma renin activity of 12 ng/ml per hour or more and an absolute plasma renin activity increase of 10 ng/ml per hour or more, along with a 150 percent increase in plasma renin activity (or a 400 percent increase if the baseline plasma renin activity was below 3 ng/ml per hour). Retrospectively, the test identified, among 200 hypertensive patients without evidence of renal dysfunction, all 56 patients with proved renovascular disease. In this group, false-positive results occurred only in two of 112 patients with essential hypertension and in six with secondary hypertension. Nine untreated patients had blood pressure levels of less than 160/100 mm Hg. The test was neither as sensitive nor specific in the 46 patients with renal insufficiency. This study demonstrates that the renin response to oral captopril is a useful screening test for identifying patients with unilateral or bilateral renovascular disease. Since the test also characterizes the renin dependency of the hypertension, it may have other diagnostic and therapeutic uses.     

单侧肾动脉狭窄患者肾静脉肾素活性的变化及意义

目的探讨动脉粥样硬化性肾动脉狭窄患者肾静脉肾素活性变化及其和介入治疗后血压变化的关系。方法测定50例单侧肾动脉明显狭窄(管腔狭窄≥70%)患者肾动脉支架置入前双侧肾静脉和外周静脉肾素活性及血管紧张素Ⅱ浓度,分析其与肾动脉狭窄及支架术后血压变化的关系。结果同时伴有冠状动脉狭窄的患者成功接受冠状动脉及肾动脉血运重建术。狭窄侧肾静脉肾素平均活性明显高于对侧[狭窄侧(1.44±1.73)ng·ml-1·h-1和对照侧(1.27±1.57)ng·ml-1·h-1,P=0.04],14例(28%)狭窄侧与对侧肾素活性比值≥1.5(肾血管性高血压组)。平均随访(12±9)个月,9例术后血压转为正常,其中肾血管性高血压组7例[50%,与对照组2例(6%)比较,P<0.001]。多因素回归分析表明肾素活性比值≥1·5与肾动脉支架术后高血压治愈相关(风险比值OR=3.15,95%CI为1.49~5.97,P=0.02)。结论约三分之一动脉粥样硬化性肾动脉狭窄患者狭窄侧肾静脉肾素分泌显著增加(≥对侧1.5倍),这些患者中半数高血压在肾动脉支架术后可以治愈。     

Treatment of Neurofibromatosis Associated Renal Artery Stenosis with Hypertension by Percutaneous Transluminal Angioplasty

A 12 year old girl with severe arterial hypertension was found to have neurofibromatosis associated bilateral stenoses of the main renal arteries and elevated plasma renin activity in the right main renal vein. Antihypertensive treatment was unable to normalize blood pressure. PTA of the right renal artery from a left axillary approach resulted in normalization of blood pressure and peripheral plasma renin activity. PTA seems to be an effective and safe method for treatment even of complicated forms of renal artery stenosis.     

Intrarenal de novo production of angiotensin I in subjects with renal artery stenosis

To estimate the renal extraction and de novo production of angiotensin I and to assess the contribution of blood-borne renin to renal angiotensin I production, the aortic and renal venous plasma levels of renin and intact [125I]angiotensin I and endogenous angiotensin I during continuous systemic intravenous infusion of monoiodinated [125I]angiotensin I were measured in subjects with unilateral renal artery stenosis (n = 8) who were treated with captopril (50 mg b.i.d.). Results demonstrated that 80% of angiotensin I delivered by the renal artery was extracted both by the affected and the unaffected kidney and that on both sides a major part of angiotensin I in the renal vein was derived from intrarenal de novo production. Production of plasma angiotensin I was in excess over extraction (p less than 0.01) on the affected side, whereas extraction was in excess over production (p less than 0.01) on the contralateral side. The plasma level of de novo intrarenally produced angiotensin I in the renal vein was seven times higher on the affected side than the contralateral side. This difference was by far too big to be explained by a difference in the transit time of blood between the two kidneys, by an augmented production of angiotensin I in the circulating blood passing through the affected kidney due to the higher level of venous plasma renin activity in that kidney, or by the combination of both.(ABSTRACT TRUNCATED AT 250 WORDS)     

Twice-daily low-dose captopril in diuretic-treated hypertensives

Twice-daily captopril (25 mg) and placebo were compared in ten hypertensive patients who were already receiving bendrofluazide. After six weeks therapy, captopril produced significant antihypertensive effects one to six hours after dosing but these did not persist at eleven to twelve hours. Plasma renin concentration was increased for twelve hours after captopril but inhibition of angiotensin II activity was lost by twelve hours. During the period when captopril reduced blood pressure significantly, effective renal plasma flow and hepatic blood flow were unchanged although renal vascular resistance was reduced. There was no evidence that captopril altered plasma sodium, potassium or magnesium concentrations following bendrofluazide. Thus, in thiazide-treated patients, captopril 25 mg produces significant blood pressure reduction for at least six hours after dosing, without impairing renal or hepatic blood flow. However, twice-daily low-dose captopril does not adequately control blood pressure throughout the dosage interval.     

Responses of the stenosed and contralateral kidneys to [Sar1, Thr8] AII in human renovascular hypertension

To better define the intrarenal hemodynamic effects of angiotensin in human renovascular hypertension, 10 patients underwent renal hemodynamic and functional measurements before and during infusion of a competitive angiotensin analog, [Sar1, Thr8] AII. Eight had technically satisfactory split function studies. Despite a fall in mean arterial pressure (132 +/- 6 to 121 +/- 6 mm Hg, p less than 0.05) and humoral changes consistent with angiotensin-mediated hypertension, the intrarenal effects of this analog were commonly those of an angiotensin agonist, producing vasoconstriction and sodium retention. This was quantitatively greatest in the contralateral kidney, whose preinfusion sodium excretion (86 +/- 30 microEq/min vs 25 +/- 9 microEq/min, p less than 0.02) and glomerular filtration rate (76 +/- 7 ml/min vs 41 +/- 7 ml/min, p less than 0.01) were higher than the stenotic kidney. In some cases, an increase in renal blood flow and rise in sodium excretion were evident during angiotensin blockade, suggesting a tonic intrarenal action of angiotensin. Although renin vein renin values differed markedly between the stenotic and contralateral kidney (ratio = 2.05 +/- 0.30), relative changes in effective renal plasma flow were correlated (r = 0.84: p less than 0.01) during infusion of this analog. These results underscore the differences in sensitivities between vascular beds to the effects of angiotensin II and the major role of the contralateral kidney in renal function and sodium homeostasis in human renovascular hypertension.