Revascularization for Renovascular Disease: A Scientific Statement From the American Heart Association

Abstract

Renovascular disease is a major causal factor for secondary hypertension and renal ischemic disease. However, several prospective, randomized trials for atherosclerotic disease failed to demonstrate that renal revascularization is more effective than medical therapy for most patients. These results have greatly reduced the generalized diagnostic workup and use of renal revascularization. Most guidelines and review articles emphasize the limited average improvement and fail to identify those clinical populations that do benefit from revascularization. On the basis of the clinical experience of hypertension centers, specialists have continued selective revascularization, albeit without a summary statement by a major, multidisciplinary, national organization that identifies specific populations that may benefit. In this scientific statement for health care professionals and the public-at-large, we review the strengths and weaknesses of randomized trials in revascularization and highlight (1) when referral for consideration of diagnostic workup and therapy may be warranted, (2) the evidence/rationale for these selective scenarios, (3) interventional and surgical techniques for effective revascularization, and (4) areas of research with unmet need.

This scientific statement has the overall goal of summarizing the current status and identifying knowledge gaps regarding renal revascularization for renovascular hypertension, ischemic nephropathy, and congestive heart failure. Renovascular disease refers to an occlusive vascular disease in arteries that perfuse one or both kidneys with attendant activation of pressor systems, sodium avidity, reduction in glomerular filtration rate, or all three. Therefore, clinical manifestations of renovascular disease extend across diverse fields of hypertension, chronic kidney disease, and heart disease. For many years it was axiomatic that restoring renal blood flow, either surgically or using endovascular techniques, represented an essential therapeutic goal. Results of several prospective, endovascular interventional studies have challenged the value of routine clinical revascularization for atherosclerotic disease, because adding it to current medical therapy has not demonstrably improved overall clinical outcomes. These data are discordant with numerous observational studies demonstrating that successful restoration of blood flow can reverse complications of untreated renovascular disease such as refractory hypertension or ischemic nephropathy, or both.¹ Essentially all hypertension treatment guidelines, including those in 2017 to 2018 from the American Heart Association and the European Society of Hypertension, emphasize the importance of treating secondary causes of hypertension including renovascular disease.^2,3 A Kidney Disease Improving Global Outcomes Controversies Conference in 2020 identified renal revascularization as a central priority in peripheral vascular disease.⁴ These issues have become more pressing as the goal blood pressure levels for optimal cardiovascular outcomes have continued to fall. As a result, clinicians may struggle to weigh potential risks and benefits of renovascular intervention against those of refractory hypertension or progressive kidney injury, or both for an individual patient. This scientific statement reviews the conditions for which renal revascularization may be beneficial.

DEFINITIONS

Renovascular Hypertension Versus Ischemic Nephropathy

Renovascular hypertension represents one manifestation along a spectrum of clinical syndromes associated with renovascular disease.⁵ It can be caused by multiple disorders, although most derive from atherosclerosis (Figure 1A) or fibromuscular dysplasia (FMD). Renovascular disease remains a prototype of secondary hypertension first demonstrated in the 1930s when experimental renal artery occlusion was shown to raise systemic arterial pressure.⁶ Revascularization of the kidney was shown to reverse hypertension, occasionally leading to a complete cure.⁷ It is important to note that the activation of pressor pathways associated with renovascular hypertension can occur with reduced perfusion pressure and kidney blood flow below the level that produces tissue hypoxia and tissue injury within the kidney.⁸ When the latter occur, ischemic nephropathy ensues.

Figure 1. Example of severe atherosclerotic renovascular disease with extensive aortic disease, pretreatment (A) and posttreatment with an endovascular stent (B).

Courtesy of Dr Sanjay Misra (Mayo Clinic).

Sequelae and Mechanisms of Renal Tissue Hypoxia, Kidney Injury

Activation of the renin-angiotensin-aldosterone system is fundamental to the development of renovascular hypertension. Angiotensin II receptor deficiency prevents renovascular hypertension.⁹ Renin-angiotensin-aldosterone system activation is followed by induction of multiple additional hypertensive pathways, including renal sodium retention, vascular remodeling and rarefication, endothelin release, activation of the sympathetic nervous system, oxidative stress, and mitochondrial damage among others (Figure 2).¹⁰ After some time, alternative pressor mechanisms become dominant as observed in experimental renovascular hypertension, and neither renal revascularization nor renin-angiotensin-aldosterone system blockade reduce arterial pressure consistently.¹¹ It is likely that the shift away from acute pressor mechanisms participates in the diminishing effect of revascularization to reverse hypertension over time, as noted later in this scientific statement. Although renal blood flow can be reduced 30% to 40% without inducing changes in intrarenal oxygenation, more severe and prolonged reductions induce overt tissue hypoxia, inflammatory injury, and eventually irreversible kidney fibrosis.¹² These changes reflect a transition over time from primarily a hemodynamic disorder to one of permanent parenchymal scarring (ie, ischemic nephropathy).

Figure 2. Progression of renovascular disease and the limitations of revascularization.

Central arrow depicts putative sequence of events associated with progressive vascular occlusion. Reductions in blood flow and perfusion pressure develop only after substantial lumen occlusion (>70%), leading to activation of the renin-angiotensin-aldosterone system. Bottom images depict blood oxygen level–dependent magnetic resonance slices mapping deoxyhemoglobin within the kidney. Despite reduced blood flow, renal oxygenation is significantly reduced only with severe and prolonged occlusion (bottom right), which is associated with activation of inflammatory injury, vascular rarefication, and tissue fibrosis. Renal revascularization can restore kidney perfusion and reverse these processes only under conditions that have not become permanent (depicted in red).

TYPES OF RENOVASCULAR DISEASE AND ASSOCIATED PREVALENCE

Atherosclerotic renovascular disease (ARVD) is the most common pathogenesis with luminal obstruction usually in the large arteries. ARVD can be identified in 6.8% of the population >65 years of age.^13,14 Although hypertension, in general, is more prevalent and more severe in non-Hispanic Black patients than in White patients, the rates of ARVD are not higher in Black patients than in White patients. In patients with severe hypertension, there is also a similar prevalence of ARVD among racial and ethnic groups.^15–17 There are fewer Black patients in numerous series of revascularization including the Cooperative Study.^15,18 However, similar outcomes of surgical revascularization for blood pressure response or recovery of renal function are reported.¹⁵

The prevalence of ARVD, visualized as >50% lumen stenosis, increases in patients with traditional cardiovascular risk factors (age, male sex, smoking, hypercholesterolemia, low high-density lipoprotein, and peripheral artery disease), with the highest prevalence in patients with peripheral artery disease (14%–40%).^13,19,20 Many of these lesions pose only minor hemodynamic effects until lumen occlusion approaches 70% to 80%. The prevalence of renovascular hypertension among patients with hypertension is estimated to be 0.1% to 5%.^21–23 The prevalence of ARVD also increases with worsening severity of hypertension, for example, documented as 14% to 24% of patients with resistant hypertension undergoing catheterization for suspected cardiac atherosclerosis.^24–26 In the prestatin era, ARVD commonly progressed with an increase in the degree of stenosis over time in 36% to 71% of patients.^27–31

FMD is a group of noninflammatory, nonatherosclerotic arterial diseases affecting predominantly middle-aged women.^32,33 In a US registry, the renal arteries are involved in 63% and hypertension is a presenting sign in 57% of patients.³⁴ In a European registry, recruited primarily from hypertension specialty centers, the renal arteries are involved in >90%, and hypertension is a presenting sign in 72% of patients.³³ About one-third of patients with multifocal FMD (the most common type) and 90% of patients with focal FMD warrant renovascular intervention³⁵ (Figure 3). Other, but less common, causes of renovascular disease include renal artery aneurysm, dissection, extravascular compression, infarction, mid aortic coarctation, partial or complete renal artery coverage by stent grafts, allograft inflow obstruction, and anatomic variants (eg, median arcuate ligament syndrome).

Figure 3. Examples of renal artery fibromuscular dysplasia lesions.

Multislice computed tomography angiographies (A, C, E) and digital subtraction angiographies (B, D, F) identify renal arteries with focal (A, B) and multifocal (C–F) fibromuscular dysplasia lesions. Irregularities in multifocal lesions (arrows in E and F) commonly represent diaphragms that produce sequential impedance to flow and pressure. Adapted with permission from Savard et al.³⁵ Copyright 2012 American Heart Association, Inc.

OBSERVATIONAL STUDIES

Early studies of surgical revascularization after short-term renal artery occlusion identify marked blood pressure reduction and recovery of kidney function in some individuals.³⁶ This was followed by extensive efforts to identify candidates for surgical revascularization with the goal of curing hypertension, which was difficult to predict, in particular, in subjects with ARVD (Figure 4). Endovascular therapy to treat ARVD was described in the late 1970s³⁸ and soon extended to FMD.³⁹ It has become the dominant form of renal revascularization. Although some patients experienced marked reductions in blood pressure, others had little sustained benefit. Efforts to select patients were based on demonstrable activation of the renin-angiotensin-aldosterone system or duration of blood pressure elevation‚ or both.

Figure 4. Individual changes in blood pressure reported in patients with atherosclerotic renovascular disease followed with sequential ambulatory blood pressure monitoring.

Individual changes in blood pressure varied significantly from reduction in blood pressure up to ≈61 mm Hg to an increase up to ≈40 mm Hg as shown. Dotted line depicts group mean fall in ambulatory blood pressure monitor systolic blood pressure of –14 mm Hg. Adapted with permission from Courand et al.³⁷ Copyright 2019 American Heart Association, Inc.

In recent years, several prospective observational studies investigating stent technologies identified average blood pressure reductions that ranged from 10 to 20 mm Hg (Table 1), with some much greater individual responses, as illustrated in Figure 4. As is evident from reports in Table 1, there have been widely variable definitions of blood pressure outcomes after revascularization. Recent studies have focused on standardizing conditions using ambulatory blood pressure monitoring–measured pressures, sometimes combined with measures to ensure medication adherence, such as witnessed pill administration. Additional outcomes of interest include slowing the progression of chronic kidney disease and treatment of heart failure. In some institutions, identification of candidates for revascularization is performed as a joint clinical review by multiple specialty physicians⁵³ or documentation of apparent treatment-resistant hypertension by ambulatory blood pressure monitoring, or both.³⁷ These reports emphasize the importance of both accurate assessment of vascular occlusion and the clinical manifestations and competing risks in considering revascularization.⁵² Characteristics and clinical features in patients observed to benefit from renal revascularization are summarized in Table 2 and are addressed in the following sections.

Table 1.

Outcomes of Renal Revascularization for Atherosclerotic Renovascular Disease

Author (y)	N	Inclusion criteria	Outcome	Features
Blood pressure outcomes
Courand (2019)37	72	Resistant HTN/ABPM	ABPM: SBP –14 mm Hg DBP –6 mm Hg	ABPM/multidisciplinary committee selection
Mangiacapra (2010)40	53	Resistant HTN/unilateral RVD	SBP –10 mm Hg DBP –2 mm Hg	Dopamine-induced gradient predictive of BP response
Leesar (2009)41	62	Resistant HTN/unilateral RVD	Office BP	BP responder (>20 mm Hg) predicted by translesional dopamine gradient (>20 mm Hg)
Ronden (2010)42	1552	ARVD meta-analysis	Meta-analysis: SBP –19.2 mm Hg DBP –9 mm Hg	BP fall after stent associated with high pulse pressure and diastolic BP
Caielli (2015)43	2133	ARVD meta-analysis	Reduced BP medications	Combined prospective randomized controlled trials and observational data: lower BP and medical Rx in stent Rx
Renal functional outcomes
Vassallo (2018)44	263	ARVD registry: Bilateral stenosis >70% Rapid decline in eGFR	Reduced end-stage kidney disease progression HR=0.35 with <1 g proteinuria	Long-term ARVD registry with 127 high-risk and 136 with no high-risk features
Ma (2016)45	253	Solitary kidney: meta-analysis	Meta-analysis: 77% improved or stable GFR	Limited to solitary functioning kidneys
Watson (2000)46	25	Creatinine >1.5 mg/dL Bilateral RAS/solitary kidney	Reciprocal creatinine shifted from negative to positive over 6 mo in 18/25 patients	Pre-Rx: slope of 1/creatinine was negative in all subjects: well-preserved kidney size
Herrmann (2016)47	30	Unilateral RAS	Total GFR unchanged Increase in stenotic kidney GFR Fall in contralateral GFR	No change in total GFR related to compensatory contralateral changes
Modrall (2017)48	60	Prestent creatinine >1.5 mg/dL Responder >20% increase eGFR	27% increased eGFR Improved long-term survival in responders (mean, 56 mo)	Median eGFR pre-Rx: 34 mL/min: 47% mortality during long-term follow-up
Cardiovascular events and mortality
Murphy (2016)49	826	CORAL cohort: stratified by initial urine protein <22.5 mg/g (n=413)	Stented group: event-free Composite outcome (73% vs 59%, P=0.02) Cardiovascular death (93% vs 85%, p<0.01) Chronic kidney disease progression (91% vs 77%, P=0.03)	Post hoc subgroup analysis with 5-y outcome data
Iwashima (2018)50	139	Resistant HTN with progression	Reduced survival after stent predicted by lower eGFR	Median follow-up 5.4 y: long-term survival predicted by eGFR at 12 mo
Takahashi (2020)51	398	Resistant HTN and chronic kidney disease: long-term follow-up	For post-stent eGFR>40 mL/min: Lower all-cause mortality Lower renal replacement Rx	Long-term (up to 10 y) Mortality: SS Death Index Renal Replacement: USRDS
Meredith (2017)52	188	ARVD >70%: multivariate analysis after stent	5-y mortality: risk factors Previous myocardial infarction Left ventricular ejection fraction <35% eGFR <45 mL/min	Cumulative mortality risk in ARVD: Stent associated with improved survival with 0,1 risk, no benefit with 2,3

ABPM indicates ambulatory blood pressure monitor; ARVD, atherosclerotic renovascular disease; BP, blood pressure; CORAL, Cardiovascular Outcomes for Renal Atherosclerotic Lesions; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; HTN, hypertension; Rx, therapy; and SBP, systolic blood pressure.

Adapted and reproduced from Mishima et al¹ by permission of Oxford University Press on behalf of the American Journal of Hypertension, Ltd. Disclaimer: OUP and AJH are not responsible or in any way liable for the accuracy of the adaptation. Licensee is solely responsible for the adaptation in this work.

Table 2.

Populations and Characteristics Considered for Renal Revascularization

Clinical populations
Unilateral renal artery stenosis with characteristic syndromes (see below)
Fibromuscular dysplasia with hypertension*
High-risk clinical syndromes*
Rapidly progressive hypertension*
Rapidly declining estimated glomerular filtration rate*
Flash pulmonary edema*
Bilateral renal artery stenosis with progressive loss of renal functional mass
Single native kidney renal artery stenosis
Special populations:
Renal allograft: transplant renal artery stenosis with or without calcineurin inhibitors
Episodic, circulatory congestion with bilateral atherosclerotic renovascular disease
Progressive loss of glomerular filtration rate with occlusive atherosclerotic renovascular disease and no other kidney disease (ischemic nephropathy)
Aortic disease with renovascular protection as part of endovascular repair
Left-ventricular assist device
Radiation-induced renovascular disease with clinical syndromes
Other diseases: eg, Takayasu arteritis, extrinsic vascular compression
Pediatric patients with mid aortic syndrome or fibromuscular variants
Characteristics suggestive of clinical benefit from revascularization
Recent onset or exacerbation (<1 y) of hypertension*
Absence of proteinuria*
Identifiable activation of renin-angiotensin system*
Hyperreninemia*
With unilateral renal artery stenosis, lateralization of renal vein renin*
Younger age
Radiographic evidence of progressive renal artery occlusion
Treatment-resistant hypertension (documentation of hypertension by ambulatory blood pressure and medication adherence)
Angiotensin-dependent glomerular filtration rate

^*Further details are available in the text.

RANDOMIZED CLINICAL TRIALS FOR RENAL ARTERY ATHEROSCLEROSIS: NEGATIVE RESULTS AND THEIR LIMITATIONS

Several prospective trials limited to ARVD have failed to identify significant benefits after renal revascularization. Taken together, these clinical trials demonstrate that routine use of renal artery revascularization for ARVD is not warranted in patients with moderate renal artery disease. However, each of these trials has important limitations that have been reviewed.^54–56 An updated systematic review funded by the Agency for Health Care Research and Quality concluded that that the “strength of evidence” for selecting modalities for therapy is low.⁵⁷ These have resulted in a missed opportunity to define the appropriate patients for whom the procedure may have benefit. Nonetheless, the use of revascularization for ARVD has fallen substantially in Europe and the United States.

The major studies since 2000 are summarized in Supplemental Table S1 and have been reviewed in detail previously. In the DRASTIC study (Dutch Renal Artery Stenosis), van Jaarsveld et al⁵⁸ randomly assigned 106 patients with ARVD with normal or mildly impaired renal function to balloon angioplasty or medical therapy alone. Investigators found no difference in the primary outcome of systolic blood pressure control at 12 months. However, 44% of the patients from the medical therapy group crossed over to the intervention group.

A decade later, 2 multicenter studies compared medical therapy with stenting for ARVD: STAR (Stent Placement for Atherosclerotic Stenosis of Renal Artery)⁵⁹ and ASTRAL (Angioplasty and Stenting for Renal Artery Lesions).⁶⁰ Both studies had defined standard medical therapy arms including blood pressure control, provision of 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors for hyperlipidemia, and antiplatelet therapy, and the intervention included angioplasty with stenting.

The STAR trial⁵⁹ randomly assigned 140 patients with normal or impaired renal function (estimated creatinine clearance <80 mL/min per 1.73 m²) and stable hypertension. This trial recruited patients on the basis of only 50% stenosis. After 2 years of follow-up, there was no difference between intervention versus medical therapy alone in the primary outcome of a 20% decrease in creatinine clearance. Almost one-third of the patients in the intervention group had a stenosis <50% at the time of angiography and were not treated. It is important to note that 30% in the medical therapy group and 44% in the stent group had arterial occlusion to a small kidney or shrunken kidney‚ or both‚ although renal size <8 cm was an exclusion. Half the patients were on fewer than 3 medications for hypertension.

The ASTRAL trial⁶⁰ randomly assigned 806 patients in which there was uncertainty of the benefit of revascularization. Sixty percent of the patients had stenosis of >70%, and crossover was minimal (6%). There was no difference in the preestablished outcomes of renal function decline, renal events, cardiovascular events, or death in the general comparison and by prespecified subgroups according to severity of the stenosis or renal function. Although the ASTRAL trial was larger than previous trials, it did not answer the question of intervention in severe ARVD.⁶⁰ Patients were enrolled if there was a stenosis of >70% by noninvasive evaluation only if the physician was uncertain if there would be benefit. Thus, patients considered likely to benefit from revascularization were excluded. Fewer than half the subjects met current criteria for resistant hypertension. Moreover, 40% of the patients at angiography did not meet entry criteria for severe renal artery stenosis.

The CORAL trial (Cardiovascular Outcomes for Renal Atherosclerotic Lesions)⁶¹ enrolled 931 patients to address whether revascularization added benefit to optimized therapy with specific blockade of the renin-angiotensin-aldosterone system for ARVD with elevated blood pressure, chronic kidney disease, or both. It is important to note that there were protocol modifications to expand the inclusion criteria because of slow enrollment. There were no differences in individual primary outcomes: major cardiovascular events (myocardial infarction, stroke, hospitalization for heart failure, or death attributable to cardiovascular disease), kidney events (progressive chronic kidney disease, the provision of renal replacement therapy, or death attributable to kidney disease), or a composite outcome. The CORAL cohort had a mean renal artery stenosis by quantitative computer-assisted angiography of 67% with <50% having severe disease (>80% stenosis).⁶¹ Hypertension was not required, and entry systolic blood pressure was 150 mm Hg on 2 antihypertensive agents with >25% already at goal. The composite end point was carefully selected to reflect the clinical events related to hypertension and progressive kidney disease, which include death attributable to cardiovascular or kidney causes, stroke, or myocardial infarction. The authors concluded that renal artery stenting “did not confer a significant benefit with respect to prevention of clinical events.” A subsequent post hoc analysis of the CORAL data after 5 years of follow-up indicated, however, that individuals with minimal proteinuria (below the median of the trial overall) experienced improved event-free survival from the composite end point (73% versus 59%, P<0.02), cardiovascular death, progressive renal insufficiency, and overall survival at 5 years (89% versus 76%, P<0.01).⁴⁹

Considered in aggregate, these trials indicate that most patients with atherosclerotic disease do not benefit from vascular intervention when treated with optimal medical therapy. Most importantly, these data do not clarify the criteria to identify people with clinical conditions and hemodynamically severe renal artery occlusive disease who may benefit from revascularization.

PATIENT POPULATIONS AND CHARACTERISTICS THAT WARRANT CONSIDERATION FOR REVASCULARIZATION

Fibromuscular Dysplasia

Hypertension is diagnosed at a similar age in patients with FMD and in patients with essential hypertension.³² FMD-related renovascular hypertension (Figure 3) should be suspected in those patients (particularly women) with early-onset, accelerated, malignant, or resistant hypertension, a small kidney without uropathy, arterial bruit in the abdomen, flank, or neck‚ or FMD in another vascular territory (Table 2).^32,62 The prevalence may be as high as 7.5% among hypertensive women <50 years of age,⁶³ and computed tomographic angiography is the initial imaging modality of choice for diagnosis of renal FMD.⁶⁴

When FMD-related renovascular hypertension is suspected, catheter-based angiography with hemodynamic assessment is warranted to determine the need for angioplasty and assess for gradient obliteration after angioplasty⁶⁴ (Supplemental Figure S1). Angioplasty without stenting is the recommended procedure when revascularization is indicated, and a consensus-based protocol for catheter-based angiography and angioplasty has been recently published.⁶⁴ Randomized controlled trials of revascularization for hypertension in patients with FMD are not available. However, revascularization has been associated with cure of hypertension in observational studies. In a meta-analysis, cure rates, defined as <140/90 mm Hg without treatment, were 36% and 54% in 47 angioplasty and 23 surgery studies, respectively. Among those who underwent angioplasty, the probability of cure significantly decreased with increasing age and duration of hypertension.⁶⁵ Also, angioplasty may be more effective in patients with focal versus multifocal FMD^35,65 (Figure 3).

Refractory Hypertension

Treatment-resistant hypertension, as defined by uncontrolled hypertension despite ≥3 hypertensive classes of medications including a diuretic or by hypertension requiring ≥4 antihypertensive classes of medications, is associated with worse cardiovascular outcomes and increased incidence of secondary hypertension.^66–68 The true prevalence of ARVD in resistant hypertension is estimated to be 14% to 23% on the basis of studies in consecutive patients with poorly controlled hypertension undergoing cardiac catheterization^24,25 and 24% in patients with resistant hypertension referred for renal arteriography.²⁶ Multiple retrospective studies of patients with ARVD and resistant hypertension who underwent percutaneous angioplasty indicate that antihypertensive pill burden can decrease (Table 1).³⁷ However, in controlled clinical trials, there are minimal differences in the number of medications in patients managed with revascularization and medical therapy versus medical therapy alone.

Progressive Kidney Function Decline

Progressive kidney function decline associated with severe hypertension and ARVD is common, but the incidence of potentially preventable ischemic renal disease is unknown.⁶⁹ In patients with existing renovascular disease, risk factors for progressive loss of kidney function or renal artery occlusion include severely elevated systolic blood pressure, diabetes, elevated peak systolic velocity >400 cm/s, and renal cortical diastolic velocity ≤5 cm/s.^31,70 Bilateral ARVD is present in 11% to 22% of patients initiating hemodialysis,^13,69,71,72 and recovery of renal function in a small subset of patients undergoing revascularization has been reported.^36,71,73 Most reports of salvage of renal function apply to patients with renovascular disease affecting the total functional mass (eg, a solitary functioning kidney or high-grade bilateral disease).^46,74 Some patients continue to have relative preservation of renal size, despite high-grade vascular occlusion. When measured kidney volume exceeds expectation on the basis of the glomerular filtration rate, the kidney has occasionally been considered hibernating and potentially salvageable.⁷⁵ It is likely that some preservation of volume is related to collateral vessels that can develop to replace a minimum level of perfusion.⁷⁶

Congestive Heart Failure

A myriad of mechanisms potentially link severe ARVD to clinical heart failure syndromes including hypertension, salt and water retention, sympathetic nervous system activation, and the effects of renin-angiotensin-aldosterone system activation on arterial stiffening and ventricular hypertrophy.⁷⁷ In addition, many people with ARVD have concomitant coronary artery disease and systolic dysfunction. Thus, it is not surprising that ARVD is common in people with congestive heart failure. In a systematic review of the literature, the prevalence was estimated at 54%.¹³

Pickering⁷⁸ described a syndrome of flash pulmonary edema in patients with ARVD. Subsequent authors reported several case series of patients with pulmonary edema who also had ARVD, and the edema improved after renal artery revascularization.^79,81 These cohorts demonstrate that 75% of patients undergoing revascularization had no further episodes of pulmonary edema after treatment.^81,83 Furthermore, Ritchie et al⁸² reported an observational series of patients with ARVD presenting with flash pulmonary edema and observed that revascularization was associated with lower mortality, but rates of cardiovascular events or end-stage kidney disease did not improve.

On the basis of expert opinion and observational data, recurrent congestive heart failure and pulmonary edema are considered indications for stenting ARVD.⁸³ This contrasts with randomized trials demonstrating that renal artery stenting provides no incremental benefit beyond optimal medical therapy for most people with ARVD without heart failure. Further high-quality data are needed to determine whether stenting adds to optimal medical therapy for patients with ARVD and heart failure. Patients with heart failure were excluded from ASTRAL and CORAL if they had been admitted for congestive heart failure within the month before enrollment (Supplemental Table S1).

CHARACTERISTICS SUGGESTIVE OF CLINICAL BENEFIT FROM REVASCULARIZATION

New-Onset Hypertension

In preclinical experiments and observation studies, a shorter duration of hypertension portends a reduction in blood pressure with restoration of blood flow. In the Cooperative Study of Renovascular Hypertension,¹⁸ analysis of 312 patients with renal revascularization for ARVD or FMD demonstrated that patients with cure or improvement of blood pressure had a duration of hypertension ≈2 years shorter than patients with a failed procedure or death (Table 2). A duration of hypertension <1 year was associated with a nearly 2-fold higher rate of cure or improvement compared with failure or death. Kirkendall et al⁸⁴ had similar findings, albeit with a smaller sample size. In a single-center observation study of 110 patients with ARVD or FMD, Hughes et al⁸⁵ showed that a duration of hypertension <5 years was associated with a significantly higher rate of cure (88% versus 30%, P<0.001). This association increased further with lateralization of the renal vein renin ratio ≥1.4 to 95%. These studies were conducted with a higher blood pressure goal (<140/90 mm Hg) and used surgical revascularization. In the 4 randomized clinical trials of revascularization (Supplemental Table S1), approximately one-third of patients in DRASTIC had hypertension for <2 years, but no post hoc analysis was available,⁵⁸ and duration of hypertension was not specified at entry in other studies.^59–61

Nonproteinuric Hypertension With Unilateral Renal Artery Disease

Proteinuria can be an irreversible sign of kidney damage that may attenuate the benefits of revascularization. In 139 patients with hypertension and renal angioplasty, Iwashima et al⁵⁰ showed that worse albuminuria was significantly associated with faster decline in estimated glomerular filtration rate. Murphy et al⁴⁹ demonstrated in a post hoc analysis of the CORAL trial that, in patients with a urine albumin-to-creatinine ratio below the median, 0.225 g/g (n=413 patients), revascularization was associated with significantly better event-free survival from the primary composite end point (73% versus 59% at 5 years; P=0.02), cardiovascular disease–related death (93% versus 85%; P≤0.01), progressive renal insufficiency (91% versus 77%; P=0.03), and overall survival (89% versus 76%; P≤0.01).

Activation of Renin

Measurement of plasma renin activity or renal vein levels of renin activity had been commonly performed during evaluation for surgical revascularization of both atherosclerotic and FMD-related renovascular disease. Early studies indicated that significant blood pressure reduction or cure of hypertension was more consistent if these studies showed lateralization of renin activity (stenotic:contralateral ratio, 1.55–1.80^86,87). When combined with known short duration of hypertension, lateralized values were reported to have high sensitivity (as high as 95%).^84,85,88 More recent studies of ARVD⁸⁹ did not show improvement, but referral criteria were more loose, and the effects of renin-suppressing drugs and perhaps, most importantly, duration of hypertension were not systematically reported. American College of Cardiology/American Heart Association appropriate use guidelines state that there is insufficient evidence to consider peripheral values of renin activity or renal vein renin ratio alone as a diagnostic procedure for ARVD.⁹⁰ Observational studies indicate that renal vein renin ratio may be most useful: (1) when considering nephrectomy of a pressor kidney, which was more common when these tests were being performed; (2) when indexed to hypertension; and (3) when imaging studies yield ambiguous results.

TECHNICAL CONSIDERATIONS FOR RENAL REVASCULARIZATION

Percutaneous Angioplasty and Stenting

Recent reports indicate that successful revascularization can be achieved in nearly 100% of subjects, although the approach and stent diameter may vary. A femoral artery approach is conventional, but radial or brachial artery access are alternatives.⁹¹ The initial technology used was 0.035-inch, and now small-diameter renal arteries using 0.014-inch platforms are being deployed with lower complication rates.⁹² Recent consensus guidelines recommend establishing the presence of a hemodynamically significant stenosis, defined as >70% lumen occlusion or mean gradient exceeding 10 mm Hg across the lesion or a fractional flow reserve <0.8 before revascularization.⁹³ Stenotic FMD lesions may be treated with balloon angioplasty alone.⁹⁴ ARVD lesions may be treated with angioplasty and stent placement, because both primary patency and restenosis are more favorable with stenting.

Potential Complications

Potential complications occurring during renal artery stent placement include atheroemboli, dissection, renal artery rupture, and thrombosis. These complications occur in 3% to 5% of patients.⁹² Complication rates have decreased with the use of lower-profile renal artery stent delivery systems. Some studies have demonstrated that atheroemboli can be prevented by using embolic protection devices. These devices have been used for coronary and carotid artery application but have been used off label for treating renal artery stenosis.^95–98 There are few data that compare technical outcomes for different platforms for renal artery interventions. Restenosis can occur, with rates between 13% and 39% by duplex ultrasound, often within a year. Many of these are asymptomatic.⁹⁹

Accessory Renal Arteries

Patients with early bifurcation of the anterior and posterior renal arteries with a stenosis involving the bifurcation or involving an accessory renal artery pose a special challenge.^100–102 An example of early bifurcation stenosis is shown in Supplemental Figure S2). In this situation, kissing renal artery stents can be placed through a 7F guide catheter with 2 drug-eluting coronary stents placed side-by-side as described elsewhere.^102–104 Approximately one-third of patients have an accessory renal artery.^103–108 These arteries are often smaller but can have stenoses that may cause hypertension. The normal renal artery measures 5 to 7 mm, and accessory arteries measure 2 to 3.5 mm.¹⁰⁹ Stenosis of a smaller-caliber artery is amenable to placement of a drug-eluting stent, as used in the coronary vasculature. This can be done in the setting of a separate origin that arises from the main renal artery.¹⁰⁹

Role for Renovascular Surgery

The role of open renal revascularization has changed with the widespread use of angioplasty and stenting. The number of open renal reconstructions steadily decreased from 1.3 to 0.3 per 100 000 adults between 1988 and 2009,¹¹⁰ in part, because of higher associated morbidity and mortality. Open renal reconstruction is reserved for a select group of patients with specific indications, primarily patients in whom previous endovascular procedures failed or who have complex renal lesions affecting arterial bifurcations, distal renal artery, or subsegmental branch vessels.¹¹¹ Longer dysplastic lesions and medial fibroplasia involving branches or complicated by a renal artery aneurysm are more suited for open reconstruction and may need ex vivo techniques because of prolonged renal ischemia.¹¹² Concomitant aortic procedures to treat aneurysm or occlusive disease may favor the use of open renal revascularization if clinically indicated.¹¹³ Children with hypoplastic renal lesions, including those in whom open or endovascular procedures failed or who have lesions associated with abdominal aortic coarctation, benefit from open renal revascularization.^114,115

Extrarenal Reconstructions

The primary goal of open aortic reconstruction techniques is to address all hemodynamically significant renal artery disease in a single operation (Figure 5). One exception is the patient with complex bilateral renal lesions who needs ex vivo renal artery reconstruction. Such patients are treated in 2 staged procedures. The choice of specific reconstruction takes into consideration the selection of incision, source of inflow (eg, direct or extraanatomic), the need for a conduit, and adjunctive techniques of renal preservation.

Figure 5. Example of aortobifemoral reconstruction with implantation of renal arteries into the aortic graft.

A, Schematic. B, Graft in situ. Such cases are less commonly performed than before but provide essential tools to deal with extensive aortic disease and protect kidney function. Courtesy of Dr Gustavo Oderich (UTHealth).

Direct reconstruction with aortorenal bypass has been most commonly used because of excellent inflow and lower rates of stenosis or occlusion.¹¹⁶ The proximal anastomosis is usually done in the infrarenal aorta or supra-celiac aorta. The choice of conduit includes saphenous vein, hypogastric artery, or thin-walled expanded polypropylene graft. In children and young adults, the use of saphenous vein may lead to late conduit aneurysm degeneration; therefore, the hypogastric artery has been preferred when possible.^7,115 Transaortic endarterectomy is an option in patients with bilateral focal ostial renal stenoses and has been used with concomitant aortic reconstructions. Endarterectomy is not suited for patients with dysplastic or longer segment lesions.

A variety of extraanatomic techniques have been described.¹¹⁷ The most common options are hepatorenal, splenorenal, or splanchnic-renal bypass. These reconstructions avoid the need for aortic cross-clamping and may limit the extent of the operation, making them favorable for the patient with higher clinical risk who is not a candidate for endovascular techniques. However, a reliable source of inflow is needed from the visceral or iliac arteries to avoid risk of graft thrombosis. Ex vivo renal artery reconstructions are reserved for complex renal lesions affecting the hilum or segmental arteries where longer ischemic time is anticipated.^118,119 The technique requires partial or complete transection of the renal vein for cooling of the renal parenchyma with infusion or renal perfusate.¹¹⁹ Intraoperative ultrasound assessment is recommended after any open renal reconstruction to identify technical problems that may lead to early graft thrombosis.¹⁶

RESEARCH DIRECTIONS

Basic Research

As can be inferred from the sequence of events in Figure 2, revascularization is effective clinically only within a limited context. Further studies related to the sequence of pressor mechanisms over the long term, changes in microvascular structure and function in the post-stenotic kidney, and activation of injury pathways in the post-stenotic kidney remain a high priority in this field. Whether adjunctive measures to restore the renal microvasculature or mitochondrial function in ischemic tissue will augment revascularization is an important area of research. Insights into mechanisms of FMD are also needed. Relevant for most patients with renovascular disease, we need to learn more about why patients with FMD are more able to respond to revascularization than patients with ARVD. Also, further insight into the mechanisms of aldosterone escape is particularly important for patients with renovascular disease because inhibitors of the renin-angiotensin-aldosterone system often anchor the antihypertensive regimen. In addition, the effects of calcineurin inhibitors on the renal vasculature require further study.

Translational and Clinical Research

Translational studies are needed for improved laboratory or imaging biomarkers to identify, among other parameters, sequence in the natural history of ischemic nephropathy and relative renin production non-invasively. Clinical studies are needed to identify more precisely the effects of age, race and ethnicity, duration of vascular occlusion, and duration of hypertension. These must be evaluated in the context of interactions with risk factors, including glucose intolerance, dyslipidemias, and associated comorbid disease. Improved diagnostic criteria to predict both blood pressure and renal functional outcomes after revascularization in specific patient groups should be given high priority. Studies focused on adjunctive maneuvers that may improve mitochondrial function in the ischemic kidney and the use of cell-based therapies to recover kidney function are ongoing. Last, additional analyses are needed to address how to reinterpret previous randomized trials in light of newer, more stringent blood pressure guidelines.²

CONCLUSIONS

Ultimately, blood supply to the kidney is essential for renal function. Revascularization is an important component in managing patients for whom vascular occlusion threatens kidney function or accelerates clinical hypertensive syndromes‚ or both. A clinical flowchart summarizing management strategies for renovascular disease is depicted in Figure 6. Recent studies underscore the adaptability of the kidneys to moderate reductions in blood flow and the limited clinical benefit of vascular intervention for moderate disease. For clinicians managing patients with renovascular disease with specific characteristics not addressed in randomized trials, the use of observational data and advances in revascularization techniques to effectuate therapy is reasonable. Basic and clinical/translational research is also needed for improved selection of patients for revascularization.

Figure 6. Flowchart depicting clinical progression to renal revascularization.

Most renovascular hypertension in adults can be managed to goal BP levels with stable renal function, regardless of pathogenesis. Further evaluation is warranted for individuals with fibromuscular dysplasia or children with FMD with or without mid aortic syndromes. Demonstrating failure to reach BP goal despite an adequate therapeutic trial may take weeks or months in older individuals. Observation over time is warranted to establish sufficient potential benefit to justify the cost and risk from interventional measures. Those with declining GFR or high-risk clinical syndromes, or both, may benefit from renal revascularization (see text). Exercise of clinical judgment includes consideration of comorbid and competing risks associated with other atherosclerotic disease or other conditions. Loss of BP control or clinical deterioration after successful revascularization, or both, may reflect recurrent disease (dotted line). ACE indicates angiotensin-converting enzyme; ARB, angiotensin receptor blocker; BP, blood pressure; CCB, calcium channel blocker; CHF, congestive heart failure; CKD, chronic kidney disease; CTA, computed tomography angiography; eGFR, estimated glomerular filtration rate; FMD, fibromuscular dysplasia; F/U, follow-up; GFR, glomerular filtration rate; HTN, hypertension; MRA, magnetic resonance angiography; PTRA, percutaneous transluminal renal angioplasty; RVD, renovascular disease; and Rx, therapy. Modified from Textor SC¹¹ with permission. Copyright 2019 Elsevier.