Hypertension
High blood pressure, if unrecognized or untreated, significantly increases the morbidity and mortality associated with coronary disease, heart failure, renal failure, and stroke. Risk further increases dramatically in the presence of smoking, glucose intolerance, hyperlipidemia, left ventricular hypertrophy (LVH), male gender, African-American race, or increasing age. Treatment of hypertension—even if only partial—can greatly reduce its morbidity and mortality risks.
Upon encountering blood pressure elevation, the first priority for the primary physician is to confirm the diagnosis. After confirmation, the evaluation focuses on three major tasks. The first is to rule out any secondary causes. Although about 95% of patients have primary disease (no clearly definable underlying cause), a search for a secondary etiology is important, because if such a cause is present, treatment will need to be etiologic to be effective. The second is to assess the severity of disease, because risk and type of treatment program derive from the degree of pressure elevation and amount of target-organ (end-organ) damage. The third is to identify any concurrent cardiovascular risk factors, because their presence will affect the threshold for initiating therapy and the nature of the treatment program. As of yet, attempts to determine the principal underlying pathophysiology have proven elusive, though when it becomes feasible to do so, the results should facilitate diagnosis and further rationalize treatment.
Definition. The definition of hypertension is arbitrary (it even varies from country to country). Actuarial data have shown that morbidity and mortality related to complications of hypertension increase linearly with increasing levels of either systolic (SBP) or diastolic blood pressure (DBP). Hence, no critical level of blood pressure exists beyond which risk becomes highly magnified. Most definitions of hypertension refer to a level of blood pressure associated with a substantial risk of complications. A major consensus report, the Sixth Report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure (JNC VI) recommends a SBP greater than or equal to 140 mm Hg and a DBP of 90 mm Hg or more for the definition of hypertension.
The definition of hypertension must be individualized for each patient. The diagnosis derives not only from the absolute level of blood pressure but also from the presence or absence of other cardiovascular risk factors. Factors besides hypertension identified by the Framingham study as significant contributors to cardiovascular risk include cigarette smoking, elevated serum cholesterol, glucose intolerance, and electrocardiographic evidence of LVH with strain. In addition, African-American race, male gender, and age greater than 50 need to be taken into account. The patient with borderline hypertension, a moderately elevated serum cholesterol level, and a history of smoking has a fivefold higher risk of incurring cardiovascular disease than the patient with borderline hypertension alone.
Classification. The JNC VI recommendations include eliminating the traditional designations of “mild,” “moderate,” and “severe” hypertension to avoid the misleading notion than mild hypertension is not a significant health risk. Instead, they designate three stages:
In addition, a “high-normal” category (DBP 85 to 89, SBP 135 to 139) is designated to highlight the increased risk of developing sustained hypertension in this group.
PATHOPHYSIOLOGY AND CLINICAL PRESENTATION
Control of blood pressure and the pathophysiology of hypertension are still incompletely understood. It has become increasingly clear that hypertension is a polygenic disorder with probable variable penetrance and phenotype in which environment may play a modifying role. Thus, it represents a complex interaction of multiple genetic and environmental factors playing varyingly significant roles in particular patients. Because there is a strong familial predisposition to hypertension, much of the pathophysiology is likely to be an expression of inherited defects in the regulation of blood pressure. There are probably several mechanistic subtypes of “primary” hypertension. It is also likely that several abnormal mechanisms are present in any one individual. It is unlikely that single gene mutations will be found to be responsible except in rare instances. As of yet, it is usually not possible to identify specific etiologic mechanisms in a given case. Nonetheless, several elements deserve elaboration and provide a rational basis for evaluation and therapy.
Primary determinants of blood pressure are cardiac output and peripheral resistance. Each is affected by a variety of factors, which, in turn, have multiple control points.
Sodium. Several lines of evidence continue to implicate sodium. Population studies demonstrate a relationship between high blood pressure and high sodium intake. In cultures where salt intake is low, hypertension is exceedingly rare. When members of those same cultures migrate into cultures in which salt intake is high, approximately 25% to 30% will develop hypertension. Sodium restriction, as well as use of diuretics, has long been known to reduce blood pressure in a subset of hypertensives. However, sodium intake is a poor predictor of hypertension in a given individual, suggesting that susceptibility depends on many other factors, probably both genetic and environmental. It is postulated that many hypertensives have an inherited defect in the ability of the kidney to excrete excess sodium. This leads to an increase in intravascular volume that is corrected by an as yet unidentified factor—the putative “natriuretic hormone”—that inhibits the Na+/K+-ATPase pump. The net result is an increase in intracellular sodium, which raises free intracellular calcium. The rise in intracellular calcium heightens vascular tone and elevates blood pressure. Natriuresis is effected at the cost of a higher resting pressure. Additionally, in salt-sensitive patients, a high sodium intake has been associated with higher levels of norepinephrine and an increased responsiveness to norepinephrine.
Catecholamines. Catecholamines affect blood pressure regulation both centrally via the vasomotor centers in the brain and peripherally through the action of the sympathetic nervous system. Catecholamines elevate blood pressure both by increasing peripheral resistance and increasing cardiac output. Sympathetic hyperactivity has been suggested as playing a primary role in the development of hypertension in some patients. Pheochromocytoma provides a model for secondary hypertension based on excessive catecholamines. Studies on patients with borderline hypertension have allowed clear identification of subgroups in which a defect in autonomic nervous system controls exists, resulting in excessive sympathetic and reduced parasympathetic activity. An exaggerated pressor response to external stressful stimuli has been demonstrated in some hypertensive patients and in their normotensive offspring. Also described are “hyperkinetic” hypertensives, who are generally young and present with tachycardia and elevated cardiac output. Their hypertension may reflect the interaction of an underlying predisposition and various environmental stimuli.
Renin-Angiotensin System. Renin is secreted by the juxtaglomerular apparatus in response to a number of stimuli, including a decrease in intravascular volume, decreased perfusion pressure, b-adrenergic stimulation, and hypokalemia. Renin acts on angiotensinogen (a decapeptide produced in the liver) to form angiotensin I, a substance with no known biologic activity. Angiotensin I is converted in the lung to angiotensin II by angiotensin-converting enzyme. Angiotensin II is a potent vasoconstrictor that also acts on the adrenal cortex to release aldosterone, which increases sodium and water reabsorption in the distal tubule of the nephron.
Renin production is inversely proportional to effective blood volume. Anything that increases effective blood volume suppresses renin and anything that decreases effective blood volume stimulates renin. For example, in primary hyperaldosteronism, autonomous production of the salt-retaining hormone aldosterone by an adrenal adenoma results in intravascular volume expansion and renin suppression. Conversely, in renal artery stenosis, decreased renal perfusion on the affected side is perceived by that kidney as decreased effective blood volume. In renin studies of patients with primary hypertension, about 15% have a high renin, the remainder showing normal or low levels. It is still not clear how much of a role this system plays in the pathogenesis of primary hypertension. One theory suggests that in patients with hypertension, a “normal” renin may in fact be inappropriately high due to a relative insensitivity to the adrenal cortical effects of angiotensin II. In addition, in about 50% of hypertensives with normal or elevated renin levels, there is a defect in the normal modulation of responsiveness to angiotensin II based on sodium intake, so-called nonmodulators. In these patients, the defect can be reversed by the use of angiotensin-converting enzyme inhibitors.
At present, we have no way to identify such patients, but this offers a glimpse of the potential of eventually being able to accurately diagnose the underlying etiology of hypertension in a given patient and adjust therapy specifically. Finally, there are local renin-angiotensin systems within the brain, heart, kidney, and placenta. It is possible that locally derived angiotensin II may play a significant role in the development of hypertension and in some of its consequences.
Insulin. The increased frequency of hypertension in patients with type II diabetes has stimulated search for a common mechanistic link. Elevations in serum insulin have been found capable of increasing plasma catecholamines and stimulating sodium reabsorption in the kidney, both of which are capable of raising blood pressure. In addition, insulin enhances the pressor responses to angiotensin II and serves as a potent growth factor for vascular smooth muscle, which could lead to hypertrophy and increased peripheral resistance. Insulin levels are higher in obese nondiabetic hypertensives than in their normotensive counterparts, suggesting a mechanism linking obesity with hypertension. Relative insulin resistance has also been identified in nonobese hypertensive patients and in nonhypertensive nonobese offspring of hypertensive parents, suggesting that elevated insulin levels may occur as part of a genetic defect and are not necessarily secondary to obesity.
Calcium. Increased intracellular calcium appears to increase vascular tone. Alteration in calcium binding at the cellular level may lead to increased levels of free intracellular calcium with a resultant increase in vascular tone.
Alteration of Cell Membrane Function. A variety of abnormalities in cellular sodium transport has been demonstrated to occur in some hypertensive patients. These include the Na+-Li+ countertransport system, the Na+-H+ exchange, the Na+-K+ ATPase pump and the Na+-K+-C– cotransport systems, among many others. The result of these abnormal transport systems is to increase intracellular sodium.
Primary or “essential” hypertension accounts for at least 95% of cases. Onset is usually between ages 30 and 50, except for isolated systolic hypertension, which is typically a disease of the elderly. Often, a family history of hypertension can be elicited. For 80% of patients, the onset is gradual and the severity mild. Patients with uncomplicated disease are asymptomatic. Although some patients report fatigue, headache, lightheadedness, flushing, or epistaxis, the correlation between symptoms and blood pressure is poor, except in patients with dangerous elevations in pressure. The rare syndrome of hypertensive encephalopathy occurs in the setting of malignant hypertension, where DBP rises rapidly above 130 mm Hg, accompanied by symptoms and signs of increased intracranial pressure (restlessness, confusion, somnolence, blurred vision, nausea, vomiting, blurred disc margins, retinal hemorrhages) and heart failure (dyspnea, rales, third heart sound).
Most patients remain asymptomatic unless end-organ damage develops, causing symptoms of congestive failure, renal failure, cerebrovascular insufficiency, peripheral vascular disease, or ischemic heart disease.
Labile hypertension is blood pressure that intermittently rises above the normal levels for each given age group and sex. Established hypertension has been shown to develop more commonly in such patients.
“White-coat hypertension” is a term used to describe blood pressure elevations that occur in the doctor’s office but not in the home or work environment. The condition is seen in both hypertensive and normotensive patients. Persons who manifest this condition typically have SBP and DBP at least 10 mm Hg greater in the office than at home but do not have greater blood pressure reactivity in the ambulatory setting. They tend to be older and are more likely to receive more antihypertensive medication than their peers, because they seem to be refractory.
Pseudohypertension occurs in elderly persons with very stiff brachial arteries secondary to fibrosis and atherosclerotic change. The vessel walls resist compression by the blood pressure cuff, resulting in sphygmomanometer readings for systolic pressure that markedly exceed the true intraarterial pressure and may simulate very high levels of hypertension. Suggestive clinical findings that differentiate this form of pressure elevation from true systolic hypertension of the elderly include absence of target organ changes (no signs of retinopathy, ventricular hypertrophy, nephropathy) despite the marked elevation in blood pressure. Osler’s maneuver (inflating the cuff above the measured SBP and seeing if a nonpulsatile radial artery can be palpated) is purported to be helpful in confirming the condition, but its efficacy is unproven.
Pseudorefractory hypertension, a form of apparently refractory disease, has been described in patients who manifest a marked vasoconstrictor response to blood pressure determinations performed with an arm cuff. Their predominant elevation is in DBP, compared with the white-coat hypertensive, who responds with a rise in SBP. Such patients are apt to be mistaken for truly refractory hypertensives because pressures may remain elevated both in the office and at home. The tipoff to this condition is the absence of end-organ damage (e.g., normal fundi, normal cardiac ultrasound) despite apparent persistence of hypertension.
Secondary hypertension has a definable etiology, occurs within a wide age range, and is often abrupt in onset and severe in magnitude; family history is commonly negative. Certain forms of secondary hypertension may be heralded by specific symptoms. For example, leg claudication may be a manifestation of coarctation of the aorta, causing lower extremity ischemia. The patient with Cushing’s syndrome may complain of hirsutism or easy bruising. Almost all patients with pheochromocytoma report paroxysms of excessive perspiration, headaches, or palpitations; about half have sustained hypertension in addition to the paroxysmal symptoms. Hypokalemia ensues from primary aldosteronism and may trigger muscle cramps, weakness, and polyuria.
About 95% of new hypertensive patients encountered in primary care practice have primary or essential disease. Secondary causes account for the remainder, with one large study showing renal failure accounting for 2.4%, renovascular disease for 1.0%, primary aldosteronism for 1.0%, drugs for 0.8%, pheochromocytoma for 0.2%, and Cushing’s syndrome for 0.1%. Coarctation of the aorta is usually detected earlier in life and rarely presents as unexplained adult-onset hypertension.
WORKUP
As noted earlier, the goals of the evaluation include firmly establishing the diagnosis; ruling out secondary causes; and determining the severity of the pressure elevation, the degree of target-organ damage, and the degree of overall cardiovascular risk.
Use of proper technique for measurement of the blood pressure is essential. Except in patients with severely elevated blood pressure, the diagnosis of hypertension should almost always be based on multiple determinations of blood pressure, preferably not only on different visits but by different personnel and in different settings. As noted above, there is a tendency for blood pressures to be higher when taken by a physician than when taken by a nurse or other medical worker. Repeating the blood pressure at the end of the visit can also be informative, because pressures are likely to be less elevated at the end of a visit than at the beginning. Studies comparing the correlation of LVH with pressures obtained in the physician’s office, at home, and at work show that work-site readings correlate best with degree of LDH.
Teaching the patient to check one’s pressure at home and at work can greatly facilitate diagnosis and management, but home determinations should be viewed as an adjunct, not as a replacement for office-based measurements. Home determinations are diagnostically useful when there is concern the office reading might represent “white-coat” hypertension caused by patient anxiety. Home determinations are usually lower than those obtained in the office. Readings in excess of 138/85 mm Hg are considered elevated. If home determinations are to be undertaken, the patient’s technique and equipment should be checked and calibrated during an office visit, comparing their readings taken in the office with those obtained by the physician using a mercury bulb manometer. In general, the mechanical aneroid manometers are simple, inexpensive, and accurate but need to be checked frequently. Finger monitors are convenient and easy to use, but they are not accurate and not recommended. If home monitoring proves to be sufficiently accurate, one can consider using such determinations to facilitate management.
When there is a marked discrepancy between home and office pressures or a wide variation in pressures obtained throughout the day, 24-hour ambulatory monitoring may be useful, though usually it is unnecessary and quite expensive. Such monitoring appears to be the most accurate way to assess blood pressure, with no regression to the mean and better correlation with left ventricular mass than casual blood pressures. Under study conditions, patients who undergo ambulatory monitoring and adjustment of therapy based on its findings need less intensive therapy, but costs are not reduced due to the expense of the monitoring. It is still not clear how to best use this information in a cost-effective manner other than to resolve discrepancies in readings if they occur.
The basic diagnostic evaluation consists of ruling out secondary causes and determining the severity of the pressure elevation, the degree of target-organ damage, and the presence of cardiovascular risk factors.
History. It is best to begin by eliciting the patient’s hypertensive history: date of onset or last previously normal blood pressure, level at time of onset, any medications taken, and response to therapy. Such information facilitates determination of etiology and helps guide workup. For example, sudden onset at a young age, very high pressure, no family history, and refractoriness to treatment suggest a secondary cause (see below). In addition, the history is checked for contributing factors, such as prior renal disease, salt and alcohol excess, and recent weight gain. Noting any associated modifiable cardiovascular risk factors (smoking, hypercholesterolemia, diabetes, and obesity) facilitates assessment of total coronary risk. Evidence of cardiovascular and neurologic complications such as a history of prior myocardial infarction or stroke and symptoms suggestive of angina, congestive heart failure, claudication, or transient ischemic attacks should be sought. It is important to inquire about the use of medications that may exacerbate hypertension such as amphetamines, oral contraceptives, corticosteroids, thyroid hormone (in excess), over-the-counter sympathomimetic decongestants, and, in the elderly, chronic use of nonsteroidal antiinflammatory agents. In addition, excessive alcohol use and cocaine abuse may elevate blood pressure.
Awareness of the symptoms associated with secondary etiologies is essential. Complaints such as hirsutism, easy bruising, paroxysms of palpitations and sweats, weakness, muscle cramps, and leg claudication should all suggest a secondary form of hypertension. Other clues to a secondary cause—especially renovascular disease—are onset at the extremes of age, rapid and severe course, and refractoriness to medication (see below).
Physical Examination. Coffee intake and smoking should be halted at least 30 minutes before taking the pressure. Blood pressure is properly measured in both arms while the patient is seated comfortably and after resting for 5 minutes. The cuff should be placed on the bared upper arm, which is supported by the examiner at heart level. The Korotkoff sounds are best listened for by using the stethoscope bell rather than the diaphragm; the bell better transmits the low-pitched sounds of diastole. The average of two successive measurements in each arm is recorded. Diastolic pressure is taken at the point at which sound disappears (Korotkoff 5) rather than when it changes in quality (Korotkoff 4). Cuff size must be adequate to avoid falsely elevated readings (cuff width greater than two thirds of arm width, length of inflatable portion greater than two thirds of arm circumference). In the elderly, the pressure should also be taken standing to detect any postural changes. Any auscultatory gap (loss and reappearance of the Korotkoff sounds) should be noted because it correlates with arterial stiffness and carotid atherosclerosis, known predictors of prognosis.
The remainder of the examination focuses on weight and pulse measurements; the skin for stigmata of Cushing’s syndrome, chronic renal failure, or neurofibromatosis; fundoscopy for arteriolar narrowing, increased vascular tortuosity, arteriovenous nicking, or hemorrhages; the thyroid for enlargement or nodularity; carotid pulses for bruits or diminution of pulse; lungs for signs of heart failure; heart for left ventricular lift and S4 and S3 heart sounds; peripheral vasculature for pulses, bruits, and abnormalities in bilateral arm and leg pressure measurements and simultaneous radial and femoral pulse palpation; abdomen for masses and bruits; and the neurologic exam for focal deficits.
Basic Laboratory Studies. The laboratory evaluation of high blood pressure has three purposes: to ascertain the degree of end-organ damage resulting from hypertension, to identify patients at high risk for the development of cardiovascular complications, and to screen for secondary possibly reversible forms of the disease. Despite the wide array of sophisticated diagnostic techniques now readily available, there is increasing evidence that the diagnosis of secondary hypertension can be made accurately and economically by the alert physician on the basis of a careful history, a physical examination, and only a few simple diagnostic tests. Extensive laboratory evaluation of patients with high blood pressure is unwarranted.
The basic laboratory investigation of hypertension for detection of secondary causes and end-organ damage and for determination of cardiovascular risk needs to include little more than a complete blood count, urinalysis, blood urea nitrogen, creatinine, potassium, calcium (with albumin), fasting blood sugar, total and high-density lipoprotein cholesterol, and electrocardiogram (ECG). The urinalysis provides evidence of primary renal disease. The extent of renal compromise due to renal disease or secondary to the hypertension itself is indicated by the blood urea nitrogen, creatinine, and urinalysis. Fasting blood sugar, serum cholesterol, and ECG supply data regarding cardiovascular risk and presence of left atrial enlargement and ventricular hypertrophy. Serum potassium is a valuable screening test for primary aldosteronism and should be known before diuretic therapy is instituted. Total cost of these determinations is reasonable. In most patients, evaluation can and should stop here.
More extensive routine laboratory evaluation of patients with high blood pressure has come under a great deal of criticism. The yield in the absence of clinical evidence for a secondary cause is low, and such testing is not cost effective. It was hoped that renin profiling would help identify the underlying pathophysiology in patients with primary disease, guide workup for secondary causes, and rationalize selection of therapy. However, the renin assay continues to suffer from difficulties with accuracy and reliability, except in certain research laboratories. Moreover, most hypertensives have normal renin levels, undermining the value of widespread testing. Finally, when renin profiling has been used to guide choice of therapy, benefit has been hard to demonstrate.
Echocardiography for detection of LVH has been useful in research studies, with presence of LVH associated with an increased risk of cardiovascular complications. However, its routine use rarely adds much to the assessment, except in the setting of refractory hypertension where definitive evidence of end-organ hypertrophy helps one distinguish between true and apparent refractoriness to therapy. When the need to search for LVH is less pressing (e.g., in the patient with newly encountered blood pressure elevation), the ECG can provide a reasonable though less sensitive estimate. Of interest, however, is the finding from the Framingham data that the presence of LVH on ECG confers higher risk than that found only on echocardiogram.
Patients at somewhat higher risk for secondary hypertension include those with abrupt onset (especially if female and younger than 35 years of age, without a family history of hypertension, or older than age 50 and with evidence of diffuse atherosclerosis), with severe hypertension (DBP more than 120 mm Hg), or with failure to respond to maximum medical therapy despite full compliance. Fortunately, in most patients at high risk for secondary hypertension, a specific diagnosis will be suggested by history and physical examination, supplemented by a few well-chosen laboratory studies.
Cushing’s syndrome is often heralded by its characteristic clinical features (e.g., truncal obesity, facial plethora, violaceous abdominal striae, proximal muscle thinning and weakness, “buffalo hump”), but the presentation may be more subtle. The initial test of choice is the 24-hour urinary free cortisol. A finding in excess of 250 µg/d is virtually diagnostic; a level above the upper limit of normal (65 µg/d) in a person with characteristic clinical features strongly supports the diagnosis, but a reading below it rules out the condition. Persons with suspected disease need an assessment of corticotropin (ACTH) dependence, which can be performed by simultaneous late p.m. determinations of plasma ACTH and cortisol. An elevated cortisol and an inappropriately normal or elevated ACTH indicates an ACTH-producing source; an elevated cortisol and a suppressed ACTH level (less than 5 pg/mL) suggests an autonomous adrenal or ectopic source. Alternatively, an overnight 1 mg dexamethasone suppression test can be performed (1 mg is taken at midnight and an 8 a.m. plasma cortisol is obtained). A cortisol level more than 5 µg/dL is suggestive of an autonomous gland, but false-positives are common (due to obesity, stress, depression, alcohol excess).
Coarctation of the aorta is suggested by the presence of reduced pulses in the lower extremities in a person with elevated arm pressures. The suspicion is enhanced by finding reduced blood pressure measurements in the lower extremities and a delay in pulse transmission on simultaneous palpation of radial and femoral pulses. In severe cases, a flow murmur may be audible over the anterior chest or back. A chest radiograph may show rib notching. Confirmation can be obtained by echocardiography or chest computed tomography (CT).
Pheochromocytoma is suggested by a story of paroxysmal sympathetic discharge (sweats, palpitations, tachycardia). The 24-hour urine assay for catecholamines, VMA, or metanephrines is a sensitive screening test for pheochromocytoma, especially when collected the same day that the patient reports symptoms and processed using the more sensitive liquid chromatography assay. Two normal 24-hour urines performed while the patient is symptomatic virtually rules out the diagnosis. Two positive urines have a high predictive value for the presence of pheochromocytoma. Methyldopa can falsely elevate metanephrines. Although combination studies are often ordered (i.e., urinary catecholamines plus metanephrines or VMA), they add little to the power of testing because their sensitivities and specificities are nearly identical. Plasma determinations of catecholamines and metanephrines are available and touted as more sensitive and specific, but sensitivity and specificity are not yet well established. If urinary screening for pheochromocytoma is positive, then one can proceed to CT of the adrenal glands—sensitivity is about 90% for lesions larger than 1 cm in diameter. Use of the CT should be limited to patients whose urines test positive and should never be used as a screening test for pheochromocytoma, because innocent adrenal masses having nothing to do with hypertension are common.
Primary hyperaldosteronism is usually suggested by otherwise unexplained hypokalemia or excessive potassium requirements in a person taking diuretics. Untreated hypertensive patients with primary hyperaldosteronism demonstrate an inappropriately high ratio (more than 20) of plasma aldosterone to plasma renin. Measuring the ratio is a reasonable screening test for suspected primary hyperaldosteronism. Confirmation requires measurements of aldosterone secretion in the setting of sodium loading and sodium depletion, which are best performed by referral to an endocrinologist.
Renovascular hypertension should be considered when blood pressure elevation is of new onset, hard to control, or associated with worsening renal function. The definitive test remains renal arteriography in combination with renal-vein renin determinations. Angiographic study with renin sampling not only identifies the anatomic and physiologic impairments, but also allows for immediate therapeutic angioplasty. However, the study has several disadvantages, including its invasiveness, risk of embolic complications, high expense, and risk of dye-induced renal injury. These risks necessitate careful case selectionn. The best noninvasive approach(es) to screening for hemodynamically significant renal artery stenosis continue to be explored.
A potentially useful clinical prediction rule was developed for screening those with drug-resistant disease or rising serum creatinine during ACE-inhibitor therapy. Using multivariate analysis, key clinical features predictive of renal artery stenosis in such patients were identified. Sensitivity and specificity for this prediction rule are about 70% and 90% respectively.
The captopril renal scan (using radionuclide scintigraphy performed 1 hour after a dose of captopril) is another means of screening. Captopril enhances differences in glomerular filtration between normal and hypoperfused kidneys. While the test shows adequate specificity (.90 to .95), it is expensive and its sensitivity is disappointing (.6 to .7). Sensitivity is especially impaired by the scan’s nondiagnostic readings in the setting of bilateral stenosis and stenosis-related complications, such as azotemia or a small poorly functioning kidney. Magnetic resonance angiography (MRA) offers a more sensitive and very specific contrast-free noninvasive means of identifying anatomically significant renal artery stenosis (sensitivity approaches 90%, specificity over 95%); however, the test is extremely expensive and, unlike captopril scanning, does not provide physiologic information.
The ultimate contributions to blood pressure control of prediction rules and these noninvasive studies singly or collectively remain to be determined. The literature should be followed for emerging data. Prediction rules have been used successfully to determine response to treatment. The younger the patient, the more recent the onset of hypertension, and the lower the systolic pressure, the greater the probability of response to revascularization for renal artery disease.
Incorporating the findings of the workup for cardiovascular risk factors, clinically overt cardiovascular disease, and target-organ damage into a risk profile helps to guide therapy. The most important factors predicting cardiovascular risk include the presence of diabetes mellitus, clinical cardiovascular disease, and target-organ damage. Damage to target organs (manifested by hypertensive retinopathic changes, signs of LVH with remodeling, and proteinuria or renal insufficiency) indicates significant risk of subsequent cardiovascular morbidity and mortality. Similarly, manifestations of overt cardiovascular disease (e.g., angina, claudication, congestive heart failure, stroke, carotid bruit) portend poor outcome if hypertension remains untreated. The JNC VI identifies three risk categories of increasing severity (A, B, and C) based on these determinants that can be used to guide clinical decision making: