Chapter 11 / Hypertensive Renal Disease

11

171

Diagnosis and Management of Hypertensive Renal and Renovascular Disease in the Elderly Wanpen Vongpatanasin, MD and Ronald G. Victor, MD CONTENTS INTRODUCTION RENAL PARENCHYMAL DISEASE RENOVASCULAR HYPERTENSION REFERENCES

INTRODUCTION The prevalence of both renal parenchymal and renovascular hypertension (RVH) increases sharply with age, making these the two most common causes of secondary hypertension in the elderly (1). The main causes of renal parenchymal disease in the elderly are diabetes mellitus and hypertension (2); the main contributor to RVH is atherosclerosis. Because the prognosis of patients with end-stage renal disease (ESRD) over the age of 60 years is extremely poor, tight control of blood pressure (BP) is essential in preventing renal disease progression. Antihypertensive medications that block the renin–angiotensin–aldosterone system (RAAS) appear to have renoprotective effects beyond BP lowering and should be first-line therapy. The target goal BP for elderly patients with renal parenchymal disease from diabetes mellitus (DM) or primary From: Clinical Hypertension and Vascular Diseases: Hypertension in the Elderly Edited by: L. M. Prisant © Humana Press Inc., Totowa, NJ

171

172

Hypertension in the Elderly

glomerulopathy with proteinuria greater than 1 g per day is now set at below 130/80 mmHg. The diagnosis of RVH should be considered in patients with generalized atherosclerosis, resistant hypertension, abdominal systolic/diastolic bruit, azotemia that is unexplained or induced by treatment with an angiotensin-converting enzyme (ACE) inhibitor or angiotensin-receptor blocker (ARB), recurrent flash pulmonary edema, or discrepancy in kidney sizes. Because clinical responses to both percutaneous intervention and vascular surgery for RVH are highly variable, revascularization should be considered only in patients with (a) bilateral renal artery stenosis or stenosis of the unilateral functioning kidney, (b) rapid decline in renal function, (c) resistant hypertension despite three or more antihypertensive medications, or (d) recurrent pulmonary edema.

RENAL PARENCHYMAL DISEASE Epidemiology Renal insufficiency constitutes a common cause of secondary hypertension, second only to RVH. In the United States, at least 30% of nondiabetic persons over the age of 60 years have some degree of renal impairment (3). Elderly hypertensives are more susceptible to develop renal insufficiency and ESRD than younger individuals (4,5). Of hypertensive patients older than 60 years with normal renal function, 10% develop elevated serum creatinine within 1 year, whereas only 5% of hypertensive patients aged 50 to 59 and 2% of those aged 30 ot 49 develop hypertension-related renal insufficiency (5). It is clear that ESRD has now become a geriatric disease. The mean age of patients who enter a dialysis program increased from 58 in 1990 to 61 in 1999 (6). During the d 1990s, the rate of increase in incidence of dialysis slowed in all other age groups except in the elderly (age 65 years or older), for whom the rate of rise has been exponential (Fig. 1). DM and hypertension are two major causes of ESRD in the elderly; glomerulonephritis and cystic disease of the kidney are less frequent (Fig. 2). Elderly patients who develop ESRD have a poor prognosis, with average life expectancy of only 3 years and an annual mortality rate of 34% (6). Cardiovascular disease is the most common cause of death, accounting for 40 to 60% of the mortality rate (6,7).

Pathogenesis Aging is inevitably associated with loss of nephron mass. In normotensive individuals, the number of normally functioning glomeruli decreases with age (8), and the glomerular filtration rate (GFR) declines at

Chapter 11 / Hypertensive Renal Disease

173

Fig. 1. Temporal trend in incidence of dialysis (per million population) in different age group. (From ref. 6.)

Fig. 2. Primary causes (%) of end-stage renal disease among different age populations in the United States. DM, diabetes mellitus; GN, glomerulonephritis; HTN, hypertension; PCKD, polycystic kidney disease. (From ref. 6.)

174

Hypertension in the Elderly

the average rate of 1 mL/minute/1.73 m2 per year (9). In untreated hypertensives, this rate is accelerated 10-fold (10). Renal parenchymal hypertension traditionally has been viewed as largely volume-dependent because of the failing kidney’s inability to excrete salt and water. However, in the overwhelming majority of patients, the main hemodynamic fault is increased systemic vascular resistance with an inappropriately “normal” cardiac output. This suggests either impaired vasodilator mechanisms or augmented vasoconstrictor mechanisms. Among these is activation of the RAAS. Plasma renin activity is inappropriately normal or mildly elevated despite an expanded plasma volume and a reduced nephron mass (11). The interaction of angiotensin II (Ang II) with angiotensin subtype I (AT1)receptor accelerates numerous cellular processes that contribute to hypertension. These include aldosterone release, peripheral vasoconstriction, and production of superoxide anion and reactive oxygen species that inactivate nitric oxide (NO) and impair endothelial function. Other potential pathogenetic mechanisms include overactivity of the sympathetic nervous system, caused by either activation of central AT1I receptors and/or uremic metabolites acting on excitatory renal afferents in the failing kidneys (12). Asymmetric dimethylarginine, a putative endogenous inhibitor of nitric oxide synthase normally cleared by the kidney, accumulates excessively in the plasma of patients with ESRD. The resultant NO inhibitor may contribute to both hypertension and mortality in patients with ESRD (7).

Diagnosis Evaluation of elderly patients with renal parenchymal disease should include urinalysis to detect proteinuria or hematuria. Renal sonography should be performed to exclude urinary tract obstruction, to exclude adult polycystic kidney disease, and to determine kidney size. Renal insufficiency should be considered when there is proteinuria by dipstick or when the serum creatinine level is 1.4 mg/dL or higher for hypertensive men and 1.2 mg/dL or higher for hypertensive women. However, renal function at a given serum creatinine level is usually much lower in the elderly than in younger individuals because of decreased muscle mass that accompanies aging. Thus, the diagnosis of renal insufficiency should be confirmed by demonstration of creatinine clearance below 60 mL/minute or urinary protein excretion above 200 mg/24 hours. A 24hour urine creatinine collection is less reliable than calculated creatinine clearance, using Cockcroft-Gault equations, in estimating GFR because of day-to-day variation in creatinine excretion and collection errors frequently found in this population (13). The urinary clearance of 125I-

Chapter 11 / Hypertensive Renal Disease

175

iodothalamate is considered to be a gold standard in measurement of GFR but is not readily available in most clinical facilities.

Treatment In patients with mild or moderate renal insufficiency, stringent BP control is imperative to slow the progression to ESRD because untreated hypertensives have a rate of loss in GFR that is 10 times faster than for normotensive individuals (10). ACE inhibitors have been shown to reduce the rate of decline in GFR in diabetic or nondiabetic chronic nephropathy with mild-to-moderate severity (14,15). Among African Americans with hypertensive nephrosclerosis, the ACE inhibitor ramipril was found in the African American Study of Kidney Disease (AASK) trial to be superior to both the dihydropyridine-calcium channel blocker amlodipine and the β-blocker metoprolol in preventing the decline in GFR, ESRD, and death (16). Of note, excellent and comparable control of hypertension was achieved in all three arms of the study. In hypertensive patients with type 2 diabetic nephropathy, ARBs or ACE inhibitors are now first-line therapy because of the results of three large trials (17–19). ACE inhibitors remain first-line therapy for type 1 DM (14) because there are no data on renoprotective effects of ARB-based therapy in this population. Elderly patients with systolic hypertension and mild renal insufficiency (serum creatinine 1.4–2.4 mg/dL) should be treated with thiazide diuretics because they derive even more benefit in terms of cardiovascular prevention than those with normal renal function (20). The target BP in elderly hypertensives with renal insufficiency is still controversial. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (21) recommended a BP goal of 130/80 mmHg for patients with chronic kidney diseases. The recommendation was largely based on results from post hoc analysis of the Modification of Diet in Renal Disease (MDRD) Study (22). The MDRD study demonstrated that treatment of hypertension to achieve a lower target BP goal (≤125/75 mmHg for patients 60 years of age or younger and ≤130/80 for patients 61 years or older) is associated with slower decline in GFR only in patients with proteinuria of 1 g per day or above. More recently, however, the AASK failed to demonstrate that, for patients with hypertensive renal disease, a lower BP goal (mean arterial pressure [MAP] of 92 or less or BP of 125/75 mmHg or less) was any better than the higher BP goal (MAP of 102–107 or BP of 130/85 to 140/90) in terms of renoprotection (16). The difference in the results of the two studies may be related to patients’ characteristics. Although the MDRD study patients were mainly Caucasians with primary glomerular disease and polycystic kidney disease, the

176

Hypertension in the Elderly

AASK study participants all were African Americans with hypertensive nephrosclerosis. For diabetic patients with hypertension, the Hypertension Optimal Treatment study demonstrated a dramatic cardiovascular benefit of achieving a target diastolic blood pressure goal of less than 80 mmHg (23). Taken together, the existing data support the goal BP of less than 130/80 mmHg for elderly patients with DM or primary glomerular disease with proteinuria above 1 g per day and the BP goal of less than 140/90 mmHg for those with hypertensive nephrosclerosis with nonnephrotic range proteinuria. In patients with far-advanced renal insufficiency, hypertension often becomes difficult to treat and may require either intensive medical regimen including loop diuretics, potent vasodilators, and central sympatholytics or initiation of chronic hemodialysis as the only effective way to reduce plasma volume. The target BP for chronic dialysis patients is even more controversial. Although some prospective (24) and retrospective studies (25) showed the positive correlation between BP and mortality, others showed no association (26) or even the inverse correlation (27). The largest prospective study in hemodialysis patients (28) suggested a U-shaped curve correlation between systolic blood pressure (SBP) and mortality. The cardiovascular mortality was increased when SBP was less than 110 or above 180 mmHg. The precise mechanism by which patients with BP above or below this range experience increased mortality is not known. Patients with very low SBP may have occult left ventricular dysfunction; those with high systolic pressure may have increased arterial stiffness, which predisposes to left ventricular hypertrophy and increased cardiovascular mortality (29,30). Randomized prospective studies are needed to determine if therapy that aims to improve arterial stiffness and maintain SBP in a specific range will translate into improved long-term outcome.

RENOVASCULAR HYPERTENSION Definition and Epidemiology RVH is the most common cause of secondary hypertension in the elderly, accounting for 5 to 7% of hypertension in patients over the age of 60 years (Fig. 3) (1). Atherosclerosis is the major form of renal artery pathology in the elderly because fibromuscular dysplasia is seen predominantly in young adults. The incidence of atherosclerotic RVH is increasing in the United States, reflecting the increased life expectancy and the aging of our population. According to the US Renal Data System database, the incidence of ESRD related to renovascular diseases has doubled over the past decade (from 2.9 to 6.1 per million per year), rising at a faster rate than ESRD related to such other causes as DM (31) (Fig. 4).

Chapter 11 / Hypertensive Renal Disease

177

Fig. 3. Effects of age on prevalence of renovascular hypertension (%) among patients referred to hypertension clinic. (Data adapted from ref. 1.)

Fig. 4. Adjusted incidence rate of end-stage renal disease caused by renovascular disease. (Reprinted with permission from ref. 31.)

It is estimated that 15 to 20% of elderly patients who enter a dialysis program have atherosclerotic renal artery stenosis as a contributing factor (32,33). The prevalence of RVH is thought to be lower in African Americans than Caucasians (34,35). However, this is an incorrect notion based on the captopril renogram. Angiographic studies have not confirmed such an ethnic difference (36,37). Atherosclerotic renovascular disease may lead to deterioration in renal function, ischemic nephropathy. Chronic underperfusion is thought to be the main mechanism leading to renal atrophy. Because renal dysfunction and atrophy are less common with fibromuscular dysplasia than

178

Hypertension in the Elderly

atherosclerotic renal artery stenosis, additional mechanisms such as atheroembolism or damage in contralateral kidney from long-standing hypertension (38,39) may contribute to renal dysfunction in the elderly with renovascular disease. It is important to emphasize that all patients with renal artery stenosis do not develop hypertension or renal dysfunction. Between 15 and 30% of patients undergoing abdominal aortogram or coronary angiogram were reported to have incidental renovascular disease. Only half of these patients have hypertension or renal dysfunction (40,41). One-third of elderly patients with congestive heart failure were found to have stenotic renal artery disease, but only one-third of these had hypertension (42). Thus, renal artery stenosis should not be used synonymously with RVH or ischemic nephropathy.

Natural History and Prognosis In the, elderly atherosclerotic renal artery stenosis tends to progress over time. The 3-year incidence of renal atrophy is 10 to 20% (43), doubling of serum creatinine is 15% (44), and progression to ESRD is 7–10% (44–46). Patients with more severe disease are more susceptible to have ipsilateral renal atrophy, but those with high SBP are also at increased risk of renal atrophy in both ipsilateral and contralateral kidneys independent of severity of stenosis (43). Thus, the presence of renal artery stenosis does not necessarily protect the kidney against harmful effects of systemic hypertension. Because renal function is influenced by many factors other than renal perfusion, it is usually difficult to show a correlation between the severity of the renal artery lesion with either baseline renal function or the subsequent decline in renal function (46,47). Once renal dysfunction develops, the prognosis of elderly patients with renovascular disease is poor. Death rates increase from 5% per year in those with preserved renal function to 15–20% per year in those with severe renal failure (46). Elderly patients who develop ESRD have a very poor prognosis, with a 2-year survival rate of 50% (45) and a 10year survival rate of only 5% (33). The major causes of death in these patients are myocardial infarction, stroke, and congestive heart failure (44,46), reflecting generalized atherosclerosis in the coronary, carotid, and peripheral vascular beds (48), respectively.

Pathogenesis Analogous to human unilateral RVH, animals with the two-kidney, one-clip model of Goldblatt hypertension have early high levels of plasma renin. Increased production of renin from the ischemic kidney

Chapter 11 / Hypertensive Renal Disease

179

leads to increased production of Ang II and aldosterone, causing increased BP. Exposure of the contralateral kidney to high BP over time leads to glomerular hypertrophy, hyperfiltration, and pressure natriuresis (49). Thus, the animals maintain a high-renin state with normal extracellular volume. Other factors that contribute to the development of this hypertension include Ang II-stimulated release of vasoconstrictor prostaglandins (50) and reactive oxygen species (51), resulting in impairment in endothelium-dependent vasodilation, and stimulated central sympathetic outflow by an action of Ang II in the central nervous system (52). In the late phase of Goldblatt hypertension, the contralateral kidneys develop glomerular fibrosis and irreversible renal injury. Reduction in GFR in both ipsilateral and contralateral kidneys leads to an expanded plasma volume, which suppresses plasma renin activity. Removal of the clip leads to resolution of hypertension early on. In the late stages, however, hypertension persists despite removal of the clip and resolves only after removal of the contralateral kidney, indicating that contralateral kidney damage from long-standing hypertension is important in maintenance of hypertension (53). Analogous to the clinical condition of unilateral stenosis of a solitary functioning kidney or bilateral RVH, animals with a one-kidney, oneclip model of Goldblatt hypertension have an expanded plasma volume with normal or low plasma renin levels. ACE inhibitors and ARB have minimal effect on BP in this low-renin condition (54,55).

Diagnosis Angiography is considered the gold standard for diagnosing renal artery stenosis. The minimal degree of stenosis that reduces renal perfusion in humans is not known, but in dogs a diameter stenosis above 70% is needed to decrease renal blood flow and increase the systemic arterial pressure (56). Because there are currently no clinical tests that can precisely assess the functional significance of a given stenosis, the diagnosis of RVH still relies heavily on clinical presentation expo facto and the BP response to revascularization. However, such reliance may still lead to inaccurate interpretation because lack of BP responses to revascularization may occur in some patients with RVH who develop irreversible contralateral or ipsilateral renal parenchymal injury. CAPTOPRIL RENAL SCINTIGRAPHY Renal perfusion can be assessed by radionuclide imaging study before and 1 to 2 hours after administration of oral captopril or intravenous enalaprilat. The radiopharmaceuticals commonly used in this test are

180

Hypertension in the Elderly

technetium-99m (99mTc) diethylenediaminepentaacetic acid (DTPA) and 99mTc mercaptoacetyltriglycine (MAG3). DTPA is purely filtered by the glomerulus; therefore, renal uptake of DTPA is proportional to the GFR. MAG3 is cleared mostly by the proximal tubules, and its renal uptake provides an estimate of renal plasma flow. In the presence of RVH, the uptake and clearance of radiopharmaceuticals are normal at baseline but become significantly reduced after captopril, which antagonizes the action of Ang II at the efferent arterioles, causing an acute fall in renal perfusion distal to stenosis. Meta-analysis (57) and a large-scale, single-center experience (58) indicated that the test has a low-to-moderate sensitivity of 65 to 75% with a high specificity of 90% in detecting renal artery stenosis. The test is less accurate in patients with renal insufficiency and bilateral renal artery stenosis. It is reportedly less reliable in low-renin hypertension (37), which is common in the elderly (38). In many observational studies, positive captopril renal scintigraphy is reported to be highly predictive of successful control of hypertension after revascularization with a positive predictive value of 90 to 100% (59–62). However, data from prospective randomized studies challenge this concept. One study of patients with ostial atherosclerotic renal artery stenosis and positive captopril renography showed that the hypertension control was improved in only one-half of patients undergoing renal angioplasty or stenting (63). The Dutch Renal Artery Stenosis Interventional Cooperative (DRASTIC) study (64) demonstrated that, in the group of patients who were randomized to receive angioplasty, the presence of an abnormal captopril renogram did not predict BP response over the 12 months of follow-up. There were no differences in either BP or doses of antihypertensive medication between patients with normal scintigraphy vs those with abnormal scintigraphy at entry (Fig. 5). DUPLEX DOPPLER ULTRASONOGRAPHY Detection of renal blood flow velocity by Doppler sonography is another technique often employed to detect renal artery stenosis. The abdominal aorta is usually imaged first, and the peak systolic velocities (PSVs) are measured from the origin, proximal, middle, and distal segments of each renal artery. Acceleration of velocity normally occurs at the stenotic site, and the Doppler signal distal to high-grade stenosis appears dampened with low velocity, so-called tardus and parvus. The presence of a renal artery PSV of 180 cm/second or above and the ratio of the PSV of the renal artery to the suprarenal abdominal aorta of 3.5 or higher indicates severe stenosis of 60% or greater (65). The procedure is time consuming and highly dependent on the skill of the sonographer.

Chapter 11 / Hypertensive Renal Disease

181

Fig. 5. Failure of captopril renography in predicting blood pressure (BP) response to percutaneous renal angioplasty in the DRASTIC study (64). The systolic BP, diastolic BP, and number of antihypertensive medications were identical among the group of patients with or without abnormal scan prior to intervention. (Reprinted with permission from ref. 64.)

Bowel gas and abdominal obesity are the major limiting factors for a successful study. It also has limited usefulness in diagnosis of the accessory vessel or branch vessel disease (38). Overall sensitivity and specificity of the test are approximately 80 to 90%. Duplex Doppler sonography may have both prognostic and diagnostic value. One recent study indicated that a high renal resistance index (1 – End diastolic velocity/Peak systolic velocity) × 100  80 is a reliable predictor of unsuccessful outcome after revascularization (Fig. 6) (66). A resistance index of 80 or above is indicative of irreversible renal parenchymal disease (67). The resistance index of the contralateral kidney is often even higher than that of the kidney with renal artery stenosis (39) and may also be predictive of the clinical response to revascularization. RENAL VEIN RENIN The renal vein renin test is based on the premise that ischemic kidneys produce excessive renin, and renin production from the contralateral

Fig. 6. Change in creatinine clearance (left) and number of antihypertensive medications and mean arterial pressure (right) before and 60 months after renal revascularization in patients with high resistance index (ℜ≥80) and low index (