Diagnosis and Management of Renovascular Disease and Renovascular Hypertension


  • Michael J. Bloch MD,

    1. From the Division of General Internal Medicine/Cardiology, University of Nevada School of Medicine Reno, NV;1 the Risk Reduction Center, Saint Mary's Regional Medical Center, Reno, NV;2 the Division of General Internal Medicine/Geriatrics, Medical University of South Carolina, Charleston, SC;3 and the Primary Care Program, Ralph H. Johnson VA Medical Center, Charleston, SC4
    Search for more papers by this author
  • and 1,2 Jan Basile MD 3,4

    1. From the Division of General Internal Medicine/Cardiology, University of Nevada School of Medicine Reno, NV;1 the Risk Reduction Center, Saint Mary's Regional Medical Center, Reno, NV;2 the Division of General Internal Medicine/Geriatrics, Medical University of South Carolina, Charleston, SC;3 and the Primary Care Program, Ralph H. Johnson VA Medical Center, Charleston, SC4
    Search for more papers by this author

Michael J. Bloch, MD, University of Nevada School of Medicine, 1500 East 2nd Suite, Suite 302, Reno, NV 89502
E-mail: mbloch@aol.com


Renovascular disease is a common but complex disorder, the most common causes of which are fibromuscular dysplasia and atherosclerosis. Clinically, it can present as asymptomatic renal artery stenosis, renovascular hypertension, or ischemic nephropathy. Assessing the clinical index of suspicion remains essential in determining an appropriate diagnostic strategy. For diagnosis in patients with suspected fibromuscular disease, it may be reasonable to proceed directly to renal angiography; however, for most patients with suspected atherosclerotic disease, there are a number of noninvasive tests available that can aid in decision making. The choice of the most appropriate initial test should be based on patient characteristics, clinical presentation, and local expertise. Treatment options include medical, surgical, or percutaneous approaches. Generally, in patients with fibromuscular disease, percutaneous intervention provides durable improvement or cure of hypertension. In patients with atherosclerotic disease, the data are less consistent, and there does appear to be a group of patients who will respond well to medical management alone. As technology advances, the diagnostic and treatment paradigms will continue to evolve.

Although it is a relatively common and well-described condition, there remains tremendous debate regarding the appropriate diagnosis and management of renovascular disease (RVD). This lack of consensus is largely the result of a paucity of high-quality clinical trial evidence and the inherent complexity of the condition. RVD is a complex disorder with various causes and presentations. When discussing the presentation of RVD, careful differentiation between the following 3 terms is essential. Renal artery stenosis (RAS) refers to the anatomic presence of an obstructive renal artery lesion. RAS is actually quite common; autopsy data confirm that >25% of all patients who die of cardiovascular causes have some degree of RAS, and in patients older than 70 years, the prevalence increases to >60%. In patients undergoing cardiac catheterization, up to 40% have had some degree of angiographic RAS, while up to 15% have been reported to have RAS of >70%. Renovascular hypertension (RVH) refers to hypertension that occurs as the direct physiologic result of RAS. The reported prevalence of RVH ranges from 0.5% to 5% of the general hypertensive population. Ischemic nephropathy (IN) refers to the progressive loss of renal function due at least in part to renal ischemia. Some reports indicate that as many as 15% of patients with renal replacement have angiographic evidence of global renal ischemia. Obviously, the demonstration of RAS in a patient with hypertension or renal dysfunction does not necessarily constitute RVH or IN. Clinically nonsignificant (asymptomatic or incidental) RAS is often found in patients with essential hypertension, those with renal failure of various etiologies, and patients with normal blood pressure. Unfortunately, many studies of RVD use the mere presence of RAS as an inclusion criterion, often complicating the interpretation of their results.

The most common causes of RVD are fibromuscular dysplasia (FMD) and atherosclerosis. FMD tends to present in younger women (although many cases are not diagnosed until after age 40), while atherosclerotic disease is usually seen in older patients with traditional risk factors for atherosclerosis. Unfortunately, many studies have included patients with both FMD and atherosclerosis, once again complicating the interpretation of their results. The diagnosis and management of FMD tends to be relatively straightforward. Although these patients may be quite hypertensive, renal dysfunction is rare and response to percutaneous or surgical intervention is usually favorable. In contrast, atherosclerotic disease presents a complex clinical dilemma. Patients with atherosclerotic RVD tend to be older and often have concomitant essential hypertension and chronic kidney disease. In a patient with hypertension or renal insufficiency and documented atherosclerotic RAS, it is difficult to determine the exact attributable effect of RAS on blood pressure and renal function, which makes it difficult to predict the potential response to percutaneous intervention. Given that atherosclerosis is a systemic disorder, these patients are also at high risk for other cardiovascular complications like heart failure, stroke, and myocardial infarction, and they are at higher risk for complications during or after percutaneous or surgical intervention. As our population ages and more patients survive the complications associated with cerebrovascular and cardiovascular disease, it seems certain that the prevalence of atherosclerotic RVD will increase.

Based on the discussion above, in patients with suspected RVD, the appropriate diagnostic and treatment strategy will depend on the suspected presentation (RAS, RVH, or IN) and the suspected etiology (FMD or atherosclerosis).


FMD is a vascular disease of unknown cause that can occur in multiple vascular beds. In the kidney, it usually occurs in the mid or distal portion of the renal artery, and it can also be found in smaller accessory renal arteries. FMD is more common in younger patients and in women. In addition, perhaps in response to stretch of the renal artery, RVH due to FMD may present after pregnancy or trauma. As opposed to FMD, atherosclerosis most commonly affects the proximal or ostial portion of the renal artery; in fact, many of these lesions are actually aortic plaques that impede the ostia of the renal artery. These anatomic considerations lead to important consequences for diagnosis and management. Imaging studies that best view the proximal portion of the renal artery will maintain accuracy for diagnosing atherosclerosis but will be less accurate in diagnosing FMD, which often occurs more distally. Also, percutaneous balloon angioplasty (PTRA) alone may not be as effective in ostial lesions since there can be significant elastic recoil after balloon dilation. As described below, the use of metallic stents for proximal and ostial lesions in atherosclerotic RVD has led to improvements in angiographic and clinical results.

As described in classic animal experiments by Goldblatt and the lesser-known Loesch, the development of RVH may result from either activation of the renin-angiotensin system, volume retention, or both. In animal models of unilateral disease, hypertension develops due to activation of the renin-angiotensin system in the affected kidney with subsequent pressure natriuresis in the contralateral kidney. Accordingly, these animals tend to appear relatively hypovolemic with significant systemic vasoconstriction. In the setting of bilateral disease, there is no contralateral pressure natriuresis, and volume tends to increase. This increase in intravascular volume leads to a decrease in renin-angiotensin system activation. Therefore, these animals with bilateral disease tend to appear volume overloaded with relatively normal renin levels (although still high for their volume status) In humans, the pathophysiology of RVH is much more complex. Depending on the number of kidneys involved, volume status, and use of antihypertensive medications that affect both volume status and the renin-angiotensin system, patients may present with varying degrees of volume expansion and renin-angiotensin system activation. Increased oxidative stress has also been implicated in the pathophysiology of RVH.

The pathophysiology of IN in humans is also quite complex. Given the large amount of functional renal reserve, IN generally requires global renal ischemia (ie, bilateral RAS or unilateral RAS in the setting of a single well-functioning kidney). In most patients with renal insufficiency and RAS, however, RVD is not the only pathophysiologic mechanism involved; there is usually concomitant nephrosclerosis and glomerulosclerosis from long-standing hypertension. Small vessel arteriosclerosis, spontaneous or provoked cholesterol embolization, and underlying medical renal disease (including diabetic nephropathy) are often present as well. As such, it becomes difficult to determine the degree of renal functional improvement one should expect with renal revascularization.

In patients with IN, the use of antihypertensive agents may lead to a decrease in glomerular filtration rate. This can occur through 2 mechanisms. First, any antihypertensive agent can decrease perfusion pressure to the glomerulus by decreasing afferent arteriolar pressure and flow. Second, agents that block the renin-angiotensin system also cause a relative efferent arteriolar dilatation, which allows a greater proportion of renal blood flow to bypass the glomerulus (lowering filtration fraction) because of a decrease in back pressure (similar to opening a valve on the end of a hose).

Importantly, like other large abdominal arteries the renal artery appears to have significant flow reserve such that luminal obstruction of at least 70% appears to be necessary to lead to physiologic consequences in most patients. Review of the literature on diagnosis and treatment of atherosclerotic RVD is complicated by the fact that many studies use a cutoff of >50% to define a “significant” lesion. As such, many studies include patients who likely have essential hypertension or medical renal disease and incidental RAS. Inclusion of these patients may overestimate the value of screening tests that may not be able to discriminate the degree of obstruction and may underestimate the value of intervention.


There is increasing enthusiasm among clinicians to make the diagnosis of RVD because it is considered a potentially reversible or treatable cause of hypertension and renal failure. The low prevalence of RVH in the general population of hypertensive individuals and the cost and limitations of the available screening tests, however, mean that universal screening of all hypertensive patients is not appropriate. We and others have suggested a triage approach based on the level of clinical suspicion. Table I outlines specific clinical characteristics that, when present, should increase the suspicion of potentially treatable RVH (or IN) and suggest further workup.

Table I.  Clinical Clues Suggestive of Atherosclerotic Renovascular Disease
Abrupt onset of diastolic hypertension before age 30 or after age 55
Accelerated, malignant, or severe hypertension
Hypertension refractory to multiple-drug treatment (=3 medications, including diuretic)
Moderate or severe hypertension in the presence of diffuse atherosclerosis
Presence of a systolic/diastolic epigastric bruit
Moderate to severe hypertension with unexplained renal insufficiency
Azotemia induced by an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker
Asymmetry of kidney size
Unexplained flash pulmonary edema or recurrent congestive heart failure
Elevated C-reactive protein

When choosing an appropriate diagnostic test, the following factors need to be weighed: clinical index of suspicion (pretest probability), presumed etiology of RVD (FMD vs atherosclerosis), presence or absence of renal insufficiency, risk of complications from conventional contrast angiography, and whether a patient is on antihypertensive medications that affect the renin-angiotensin system. For patients with a high clinical index of suspicion and relatively low risk of complications, it may be reasonable to proceed directly to conventional contrast angiography; this is particularly the case in patients with suspected RVH from FMD. Conventional contrast angiography has long been considered the gold standard in the diagnosis of patients with suspected RVD. While angiography does not directly give information about the functional significance of a renal artery lesion, those that obstruct at least 70% of the cross-sectional view of the lumen or lead to a significant pressure gradient are generally considered hemodynamically significant and appropriate for intervention. Of course, conventional angiography is invasive and carries a significant potential for morbidity. This is particularly true in patients with a heavy burden of atherosclerosis who are at risk for atheroembolic disease, which may occur in as many as 5% to 10% of patients undergoing conventional angiography and is a potentially irreversible cause of renal function deterioration.

Given the potential risk of conventional angiography in patients with a lower index of suspicion or higher risk for complications of angiography (including the majority of patients with presumed atherosclerosis and/or renal insufficiency), a noninvasive study is usually the most appropriate first step. As shown in Table II, noninvasive tests can be divided into those that rely on an assessment of the functional effects of a potential renal artery lesion (plasma renin activity, captopril test, captopril-augmented renal scintigraphy, renal vein renins) and those that rely on direct imaging of the renal artery anatomy (duplex ultrasound [DU], magnetic resonance angiography [MRA], and computerized tomographic angiography [CTA]).

Table II.  Commonly Employed Noninvasive Tests for the Diagnosis of Renovascular Disease
Functional TestsImaging Tests
Plasma renin activityDuplex ultrasound
Captopril plasma renin testMagnetic resonance angiography
Renal vein reninsComputed tomographic angiography
Angiotensin-converting enzyme inhibitor-augmented scintigraphy 

In general, those studies that primarily provide a functional evaluation suffer from impaired accuracy in the presence of significant renal insufficiency or bilateral disease. Moreover, to maintain sufficient accuracy, these studies usually require the discontinuation of antihypertensive medications that often affect the renin-angiotensin system. These tests are therefore most appropriate for patients with moderate clinical suspicion of FMD or suspected uncomplicated atheromatous disease with normal renal function. Given the widespread use of antihypertensive agents that block the renin-angiotensin system, the increasing incidence of renal dysfunction among patients undergoing workup for RVD, and rapid technologic improvements in imaging studies, these functional studies have generally fallen out of favor among most clinicians. In fact, in studies reporting on selected populations, captopril-augmented renal scintigraphy is the only noninvasive study that has consistently been shown to predict the response to renal artery intervention; as such, these “renal scans” should still be considered a reasonable diagnostic option in appropriate patients.

In contrast to the functional studies, noninvasive tests that rely on direct renal artery imaging maintain their accuracy even in the presence of renal insufficiency, but they tend to lose accuracy in patients with the distal, branch, or accessory RAS common in FMD. As such, these tests are most appropriate for the patient with suspected atheromatous disease and/or renal dysfunction. Perhaps their most valuable utility is their high negative predictive value for detecting atherosclerotic RVD.

Although branch or accessory lesions may be missed, a completely normal, carefully performed imaging study makes atherosclerotic main RAS highly unlikely. Which imaging study to choose in an individual patient is as much a question of local expertise as individual patient characteristics.

DU is perhaps the most widely evaluated of the imaging studies. Significant RAS is determined by comparing the peak systolic velocity in the renal artery to that in the adjacent aorta. Using conventional contrast angiography as a reference, DU has been shown to detect RAS of more than 60%, with sensitivities ranging from 84% to 98% and specificities from 90% to 98%. Unfortunately, not all lesions that are more than 60% stenotic are functionally or hemodynamically significant. Furthermore, as traditionally performed, DU has limited ability to judge the exact degree of stenosis or predict response to intervention. It is also highly operator-dependent and centers without significant experience with this modality may not be able to replicate the accuracy demonstrated in clinical trials.

Efforts are under way, however, to investigate additional DU measurements to identify which lesions may respond best to intervention. In one study, an end-diastolic velocity >150 cm/s has been demonstrated to predict lesions >80%. In another study, subjects with an elevated resistive index (>80), suggestive of distal arteriosclerosis, had little clinical improvement after percutaneous or surgical intervention. Since the main renal arteries can be difficult to identify and trace throughout their course, especially in obese patients, other innovations to traditional DU have been studied.“Indirect” Doppler evaluation, where the examination is limited to the distal segmental renal arteries, is frequently employed in certain clinical centers. A number of clinical trials, however, have questioned its accuracy, and it should generally be considered a suboptimal test. Color or power Doppler and the use of ultrasonic contrast agents are more promising but are still experimental innovations. To summarize, in experienced hands, DU remains an important noninvasive screening tool for atherosclerotic RVD, and it remains the study of choice in identifying RAS after percutaneous intervention or in a transplanted kidney.

Contemporary MRA studies use a gadolinium contrast agent to identify the vessel lumen. While gadolinium contrast agents have previously been considered to be safe for use in patients with renal insufficiency, recent case reports from Europe identifying a potential link between gadolinium administration and development of a rare kidney disease known as nephrogenic systemic fibrosis or nephrogenic fibrosing dermopathy (NSF/NFD) prompted the US Food and Drug Administration to issue a public health advisory concerning MRA on December 22, 2006. Cases of NSF/NFD have been limited to patients with chronic renal insufficiency who received large amounts of gadolinium contrast agent such as that used in MRA; it is hoped that further information will be available shortly concerning this possible safety issue. Additional limitations to MRA include the requirement for breath-holding, long scan times, risk of claustrophobia, and contraindication in patients with pacemakers, cerebral aneurysm clips, or intra-ocular metal.

CTA using multiple rotating detectors can acquire a large amount of data in a relatively short time. Most published studies of CTA use 4-detector scanners; however, in many areas of the country 16-, 32-, and 64-detector scanners are increasingly available. This should lead to even greater resolution and the ability to more easily identify distal renal artery lesions. The principle disadvantage of CTA is the need to use significant amounts of potentially nephrotoxic iodinated contrast. This makes CTA less optimal for the increasing number of patients with chronic renal insufficiency, especially the elderly, who undergo workup for RVD each year.

Published results with both CTA and MRA are encouraging. In general, reported sensitivity and specificity in identifying the presence or absence of RAS have been in excess of 90%, and a recent meta-analysis suggested that both MRA and CTA performed better than other imaging studies in identifying RAS. Unfortunately, the majority of these studies included only a small number of patients with FMD and distal or branch RAS; when such patients are included, diagnostic accuracy declines significantly. It should also be remembered that although quality software is now widely available, significant expertise is still necessary in formatting and interpreting the images. Although both MRA and CTA are accurate in defining the presence or absence of RAS, it can still be difficult to correctly predict the exact degree of stenosis found at subsequent angiography. Improvements in resolution with the use of multidetector scanners and the addition of measurements that add hemo-dynamic and functional information should continue to rapidly overcome these limitations. In the future, CTA and MRA will increasingly become the most commonly employed noninvasive studies, at least in patients with normal renal function, but the rapid pace of technologic advance threatens to outpace our ability to understand when and how to use them.


Management options for RVD include medical management, percutaneous interventions, and surgical revascularization. Percutaneous interventions include percutaneous transluminal angioplasty (PTRA) and stenting. Medical management can be used alone or in combination with percutaneous or surgical intervention. Just as with diagnosis, the most appropriate treatment strategy for individual patients with RVD depends on the cause (FMD or atherosclerosis) and the clinical presentation (incidental RAS, RVH, or IN). The goals of management are (1) control of blood pressure, (2) preservation of renal function, and (3) avoidance of complications and adverse effects of treatment.

Surgical Intervention

Surgical procedures include endarterectomy, aortorenal bypass, and extra-anatomic bypass (bypass from the celiac or mesenteric branches). Possible complications include bleeding, infection, myocardial infarction, stroke, atheroembolic disease, and acute renal failure. Reported rates of perioperative mortality range from 1% to 6% and are usually higher with concurrent repair of aortic aneurysms. Most studies of surgical revascularization have shown excellent short- and long-term patency, around 85% and 95% at 5 years, as well as improvements in blood pressure control and stabilization of renal function. Nonetheless, surgical intervention is currently reserved primarily for complex lesions not amenable to percutaneous intervention or for patients with concomitant aortic aneurysms requiring repair.

Percutaneous Intervention

Patients with FMD usually show a substantial clinical and anatomic response to PTRA without stenting. Renal artery patency is achieved in nearly 100% of patients, and in patients with RVH due to FMD, there is a substantial cure rate (attainment of normotension without medication) of approximately 40% to 50%, with an additional 40% to 50% whose blood pressure is considered improved at last follow-up. Patients with fibromedial hyperplasia, a less common form of FMD, may have less optimal results. Although restenosis leading to repeat procedures is somewhat common, given the high rate of improvement or cure and the relatively low risk of complications, PTRA is generally considered the most appropriate initial treatment strategy in FMD.

According to claims data, there were more than 18,000 percutaneous renal artery interventions performed in the Medicare population alone in 2000. This is of concern because in the setting of atherosclerotic disease, the response to percutaneous intervention is more variable and generally less favorable than that seen with FMD. Given the elastic recoil of the ostium of the renal artery after PTRA and the propensity for restenosis, stenting appears to be a major clinical advance in the setting of atherosclerotic RVD. Although no endoluminal stents have been approved for use in the renal artery, their use has become widespread. Appropriate indications for stenting include (1) poor angiographic response to routine PTRA, (2) early restenosis after PTRA, and (3) renal artery dissection during PTRA. In addition, a number of centers have begun routinely stenting all ostial lesions. The use of drug-eluting stents to minimize restenosis and distal protection devices to prevent potential atheroembolism have also shown promise in early clinical trials, but both remain investigational at the present time.

In published series of patients with presumed atherosclerotic RVH who undergo renal artery stenting, patency is achieved in at least 90% of patients. Cure of hypertension is rare; stenting often reduces but rarely eliminates the need for antihypertensive medication. Improvement in blood pressure control is seen in about 50% to 85% of patients. In recent years, much of the emphasis of percutaneous revascularization for atherosclerotic renal artery disease has shifted from blood pressure control to preservation of renal function. Many investigators have concluded that renal function can be preserved or restored with percutaneous intervention. Taken as a whole, contemporary reports of renal artery stenting on renal function show that approximately one third of patients show long-term improvement, one third show stability, and one third show deterioration; whether this represents a significant improvement over the natural history of IN remains a matter of considerable debate.

Of course, percutaneous intervention is also associated with a risk of morbidity, including hematoma, pseudoaneurysm, renal artery dissection or thrombosis, acute renal failure, and atheroembolism. It is difficult to quantify the exact incidence of these complications, but it is somewhere on the order of 10% to 20%, most of which are minor. Atheroembolic disease is an especially insidious and underappreciated complication, with as many as 44% of affected patients requiring dialysis within 6 months of diagnosis. Patients with atherosclerosis are more likely to have complications than those with FMD. In addition, the presence of extensive aortic atherosclerosis, significant cardiovascular disease, and chronic renal insufficiency increase risk of these procedures.

Medical Therapy

In patients with FMD, there is little role for medical therapy alone, although sometimes anti-hypertensive medication will be needed for residual hypertension despite adequate revascularization. In patients with atherosclerotic RVD, aggressive medical management entails not just control of blood pressure but comprehensive therapy aimed at reducing the risk of other complications of systemic atherosclerosis, including stroke and myocardial infarction. Although outcome data are lacking, most investigators and clinicians have determined that lifelong antiplatelet therapy, smoking cessation, and aggressive control of hyperglycemia is indicated in patients with atherosclerotic RVD (with or without percutaneous intervention). In addition, although it is not specifically mentioned in the National Cholesterol Education Program Guidelines, given the high incidence of concomitant undiagnosed coronary and cerebrovascular disease, atherosclerotic RVD should generally be considered a coronary heart disease risk equivalent and lead to appropriate aggressive treatment of dyslipidemia. At least one report has suggested that aggressive low-density lipoprotein lowering with statins leads to decreased progression of athero-sclerotic RAS.

For blood pressure control, use of all antihypertensive classes are appropriate, and combination therapy is often necessary. Given the pathophysiology of RVD, angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers are frequently required to get blood pressure to goal and may have unique advantages in maintaining long-term renal function in the face of concomitant medical renal disease. While contemporary data are lacking, older studies have demonstrated that a substantial number of patients with RVH can have significant improvements in blood pressure with medical therapy alone. In these studies, acute renal failure with ACE inhibitors was rare and generally reversible with discontinuation of the agent. Although it is difficult to generalize, an increase in serum creatinine of <25% and a serum potassium of <5.5 mg/dL usually does not require discontinuation of the agent. RAS, even when bilateral, should not be considered an absolute contraindication to the use of ACE inhibitors and angiotensin receptor blockers.

Medical Therapy or Percutaneous Therapy?

It is very difficult to compare data from one published report with another due to substantial and sometimes striking differences in methodology. Published studies have employed many different entry criteria, and many studies combine patients with FMD and atherosclerotic disease. Despite published recommendations, reporting definitions of response also remain troublesome; in general, patients are often classified as cured, improved, stable, or worse, but there is little standardization.

There have been 3 prospective randomized clinical trials of percutaneous versus medical intervention; unfortunately these trials also have significant shortcomings. These include small sample sizes, potential inclusion bias, use of PTRA without stenting, lack of standardized medical intervention, and high rate of crossover between groups. A recent meta-analysis of the data stated that available studies “generally had poor methodologic quality and limited applicability to current practice” and did not clearly support one treatment approach for all patients with RVD. In general, for patients initially treated with percutaneous intervention, these trials demonstrated modest improvement in blood pressure, a slightly higher risk of complications, and no substantial effect on renal function when compared with patients who were managed with medical therapy alone. Of interest, the difference in blood pressure was substantially lower than generally reported in available observational cohort studies.

Based on the poor quality of available data, the clinician faces a difficult choice in selecting an initial treatment strategy. One thing appears certain: no one therapeutic approach is appropriate for all patients. In each patient, the individual risks and benefits associated with each potential treatment needs to be considered. Percutaneous or surgical intervention should only be undertaken when there is good evidence of a potentially hemodynamically significant RAS, either a lesion >60% (some would say >70%) and/or a significant pressure gradient (most would say at least 20 mm Hg) measured across the lesion at the time of angiography.

Although the quality of the data is poor, there have been a number of clinical characteristics that have been shown to be associated with poor outcome or an increased risk of complications with percutaneous intervention, including decreased kidney size (<8 cm), advanced renal dysfunction (serum creatinine level >3.0 or 4.0 mg/dL), another potential cause of renal dysfunction (eg, diabetes, amyloidosis), longstanding hypertension or renal dysfunction, decreased left ventricular function, renal artery occlusion, high resistive index on DU, and extensive burden of aortic atherosclerosis. Some studies have suggested that patients with bilateral disease may have modestly greater improvement in blood pressure and stabilization in renal function, but may also have a higher risk of potential complications, with percutaneous intervention. An interesting recent report also indicated that patients with a higher baseline level of brain natriuretic peptide had a more favorable blood pressure response to percutaneous intervention.

The changing epidemiology of RVD also plays a role in this decision. As our population ages and mortality from cerebrovascular and cardiovascular complications of atherosclerosis decreases, clinical manifestations of RVD are occurring in older individuals. The average age in the most recently published series of renal artery intervention is 68 to 71 years, approximately 10 years older than in similar studies published a decade ago. While published reports do not show that older patients have worse outcomes with percutaneous intervention, there may be a substantial selection bias in published reports. Older patients, with more diffuse atherosclerosis, may have a greater risk of complications, and given the high incidence of cardiovascular events in patients with atherosclerotic RAS, many may not live long enough to reap the benefits of renal artery interventions.

The American Heart Association/American College of Cardiology (AHA/ACC) recently published guidelines on when to consider renal artery intervention as opposed to medical therapy alone. Based on these guidelines, other published studies, and expert opinion, it seems most appropriate to recommend initial percutaneous intervention in all patients with RVH due to suspected FMD and in patients with atherosclerotic RAS who have specific indications as shown in Table III. No matter what the initial strategy, patients with atherosclerotic RAS should receive comprehensive medical management aimed at stabilizing atherosclerotic plaque in other vascular beds, and although outcome data are not available, we believe these patients should undergo long-term careful surveillance with frequent measurement of BP and serum creatinine (monthly to quarterly) and renal sonogram and DU or other imaging modality (semiannually or yearly).

Table III.  Indications for Percutaneous Intervention in Atherosclerotic Renal Artery Stenosis
Unable to achieve reasonable blood pressure control despite appropriate medical antihypertensive therapy
Decline in renal function or size despite aggressive medical therapy
Recurrent flash pulmonary edema or exacerbations of heart failure with relatively preserved left ventricular function
Very high-grade stenosis (>95%) with high risk of occlusion
Dialysis-dependent renal failure without other identifiable cause (particularly if of recent onset)

Given the paucity of good clinical data, there is great enthusiasm that the National Institutes of Health-sponsored Cardiovascular Outcomes in Renal Atherosclerotic Lesions (CORAL) study will provide a definitive answer concerning the initial strategy of choice in atherosclerotic RVD. The CORAL study is an ongoing prospective multi-center randomized 2-arm clinical trial that will test whether stent placement plus medical therapy will provide superior clinical outcome to medical therapy alone in 1080 patients with angiographically significant (>60%) RAS and hypertension. Medical management in the CORAL study will consist not just of aggressive combination antihypertension therapy (with an angiotensin receptor blocker as initial therapy) but also use of lipid-lowering therapy, antiplatelet therapy, and aggressive control of hyperglycemia. It is our opinion, however, that no one clinical trial will provide a suitable answer for all patients, and although it is perhaps the best study that could be designed at the time, the CORAL study does have some significant shortcomings, including potential for significant inclusion bias, exclusion of patients with significant renal dysfunction (serum creatinine level ≥3.0 mg/dL), and use of survival free from cardiovascular and renal events as the primary end point rather than blood pressure control and stabilization of renal function.

Given the high prevalence of RAS in patients with atherosclerosis in other vascular beds (usually reported to be around 10%–25%), there is increasing enthusiasm to perform abdominal and nonselective renal arteriography (often known as “drive-by” angiography) at the time of cardiac catheterization. When RAS is identified, percutaneous intervention is also frequently performed, sometimes in the absence of reasonable clinical indications. The appropriateness of these procedures is hotly debated. In a patient already undergoing coronary angiography, the additional attributable risk of performing nonselective (ie, without cannulizing the renal arteries) abdominal aortography is small. Since the presence of atherosclerotic RAS provides valuable information about the potential progression and prognosis of cardiovascular and renal disease (and even mortality in some studies), in accordance with a recent science advisory from the AHA, abdominal aortography at the time of coronary angiography appears justified in patients with clinical clues suggestive of possible RAS (as outlined in Table I).

The more difficult question is whether percutaneous intervention is warranted when significant RAS is found at the time of coronary angiography. It is certainly tempting to consider intervention whenever angiographically significant RAS is demonstrated (sometimes known as the oculu-stent reflex). However, there are limited data demonstrating improved clinical outcomes in patients undergoing percutaneous renal artery interventions in this setting. Many of these lesions may actually be incidental and not hemodynamically significant, especially those with 50% to 70% stenosis. In addition, as opposed to nonselective abdominal aortography, renal artery intervention may be associated with an increased risk of atheroembolism and other complications. In the absence of clinical trials demonstrating a favorable risk-benefit ratio, we recommend reserving renal artery stenting only for patients with clear indications as outlined above. Of course, the finding of significant atherosclerotic RAS at the time of coronary angiography is a clear indication for comprehensive medical management and disease surveillance.


Given the rapid pace of technologic advance, the paucity of high-quality clinical trial data, and the broad range of patients presenting with suspected RVD, it is difficult to formulate a single diagnostic and treatment approach for all patients. Instead, clinicians must take an individualized approach in choosing an appropriate strategy for each patient. This analysis should consider not only the patient's presenting blood pressure and renal function but also the potential for morbidity from the various diagnostic and treatment procedures. Given the increasing prevalence of RVD and its important potential health consequences, further well-designed clinical trials are needed to better refine these diagnostic and treatment paradigms.