POINT/COUNTERPOINT: Renal artery occlusive disease – To treat or not to treat? ASTRAL and CORAL trials show no indication to treat percutaneously. There are still indications to treat renal artery occlusive disease.

Percutaneous treatment of renal artery occlusive disease is unnecessary and should be abandoned, except in pediatric cases.

BY GEORGE HAMILTON, M.D.

This position is supported by findings from both the ASTRAL trial (N. Engl. J. Med. 2009;361:1953-62) and the CORAL trial (N. Engl. J. Med. 2013 Nov. 18 [doi:10.1056/NEJMoa1310753]).

The ASTRAL trial, a prospective, randomized comparison of best medical therapy with and without stent angioplasty in more than 800 patients, was the largest trial to date when it began back in the 1990s. The well-known results showed no difference in time to first renal event, first vascular and cardiovascular events, and overall survival. Furthermore, there was no difference in these outcomes among patients with greater than 90% stenosis, with the exception of a possible difference in mortality, which trended toward improvement among those with high-grade stenosis.

We concluded that revascularization in the vast majority of patients is unlikely to improve hypertension control or renal function, and that renal artery stenosis is not pathophysiologically important. We also concluded that there is no point in screening for asymptomatic disease; this was back when every patient was getting screened, and treated primarily on the basis of finding a renal arterial stenosis.

Finally, we concluded that properly applied best medical therapy alone was an extremely good treatment.

Several flaws in the trial garnered extensive criticism, however, and the more rigidly designed CORAL trial was expected to address them. The findings confirmed those of the ASTRAL trial. In more than 900 patients from 88 centers, there was absolutely no benefit of intervention with respect to primary and secondary outcomes, including among those with high-grade stenosis.

We can now see on the basis of extensive level 1 evidence that when added to comprehensive, multifactorial medical therapy, intervention yielded no benefit.

So are there certain patient groups who might benefit more from intervention? Among listed indications are high-grade stenosis (which doesn’t apply any longer); short history of progressive failure (which is quite rare); ACE-induced renal failure (which is also quite rare); difficult-to-control hypertension (there really is no such thing now, except in a tiny percentage of patients); and – the least challenged indication – flash pulmonary edema. These remaining indications move our interventions into a very high risk group of patients.

The current debate is focused almost entirely on endovascular intervention, but a systematic review showed that there is long-term benefit in terms of renal function and hypertension with open procedures. Although overall there is increased mortality, this risk is minimized – and not significantly different from endovascular procedures – in those having only renal revascularization vs. those having concomitant aortic procedures. So open surgery remains a possible treatment option, indeed a recent level 1 study comparing stenting and open surgery, showed better long-term results with open surgery (J. Vasc. Surg. 2009;49:667-75). The authors concluded that surgical reconstruction remains the gold standard in treating renal artery stenosis. Although national data suggest an overall mortality of about 10%, it is much lower at specialist, high-volume centers with mortality rates similar to those of stent angioplasty.

Renal stenting is not a low-risk procedure. In all-comers the complication rates, serious complication rates, and mortality rates are significant with short-term equivalence between focused renal arterial surgery and percutaneous intervention.

Returning to the debate, are either methods of revascularization appropriate? Probably not.

Even in flash pulmonary edema, there is little evidence to support revascularization. Few papers exist suggesting a benefit of revascularization in reduction of flash pulmonary edema, but the patient numbers were small, and there was no benefit in terms of preservation of renal function.

The history of evolution and evaluation of the role of renal revascularization is remarkably similar to that of renal denervation, initially and with considerable conviction thought to be a cure for hypertension. However, when properly assessed by prospective randomized comparison there was found to be absolutely no benefit.

So, given the considerable objective evidence from two major trials and revisiting the basics of the pathophysiology of atherosclerotic renovascular disease, to expect benefit from treating the osteal component of renal artery occlusive disease is at best naive, in my opinion. There remains little clinical evidence of benefit for any indication, with the possible exceptions of ACE-induced renal failure and possibly flash pulmonary edema in the presence of bilateral renal arterial stenoses.

Dr. Hamilton is a professor at the Royal Free London Hospital, University College London, United Kingdom.

There are still indications to treat renal artery occlusive disease

BY MATTHEW A. CORRIERE, M.D.

Although renal artery revascularization has been grossly overutilized and is not indicated in the majority of patients with renal artery stenosis, I perform renal artery revascularization as part of my routine clinical practice and believe that there are many instances where revascularization should be considered, particularly when patients have severe symptoms despite aggressive medical therapy. While neither ASTRAL nor CORAL observed any benefit associated with revascularization, both have important limitations that should be kept in mind when interpreting the results of these trials.

These limitations can be broadly categorized as mismatch between indications for revascularization and clinical endpoints, selection biases favoring enrollment of patients with relatively mild symptoms, and inconsistencies between study protocols and contemporary decision-making strategies.

Given that ASTRAL’s primary outcome was change in renal function (defined by a 20% or greater reduction in the mean slope of the reciprocal of serum creatinine), it is important to remember that the inclusion criteria were renal artery stenosis with unexplained renal dysfunction or poorly controlled hypertension. Patients who had hypertension in the absence of significant renal dysfunction were therefore eligible, and 40% of the randomized participants had preserved baseline renal function (based on a serum creatinine of < 150 micromol/liter). Unlike patients with baseline renal dysfunction (which, in theory, might improve with revascularization), these patients with normal renal function who were treated with revascularization risked decline in renal function resulting from procedure-related adverse events without any real chance of renal function improvement. It would certainly be difficult to justify revascularization for the sake of renal function salvage in these patients, and their inclusion within a randomized trial with change in renal function as its primary outcome is problematic for the same reason.

ASTRAL also had an additional, somewhat unorthodox inclusion criterion: uncertainty on the part of the treating physician that the patient “definitely would have a worthwhile clinical benefit from revascularization.” Exclusion of patients considered likely to benefit from revascularization would seem to ensure a selection bias favoring the null hypothesis; this approach may also explain the large proportion of participants with relatively mild occlusive disease (40% had stenotic lesions that were < 70% in severity).

A high rate of both technical failure (12%) and adverse events (20%) associated with revascularization, asymmetric crossover between treatment groups (86 of the 110 patients who did not receive their randomized intervention were in the revascularization group), and lack of standardized protocol for medical therapy further limit the conclusions that can be drawn from the ASTRAL results.

Although this trial does not provide us with compelling evidence that renal revascularization should be abandoned for patients failing appropriate medical therapy, ASTRAL demonstrated that no benefit should be expected from nonselective use of revascularization, which can be associated with significant rates of both technical failure and major adverse events.

The CORAL trial overcame many of the design limitations for which ASTRAL drew criticism. CORAL’s primary endpoint (freedom from major adverse cardiovascular or renal events) allowed potential benefit for participants with either systolic hypertension or chronic kidney disease as their indication for treatment. Although participants with systolic hypertension as their inclusion criterion had to be on at least two antihypertensive medications, it is important to acknowledge the growing number of indications for these medications related to cardiovascular risk reduction in the setting of diabetes, heart disease, and other diagnoses that may be unrelated to any specific blood pressure target. Number of antihypertensive medications is therefore often a crude and potentially invalid indicator of hypertension severity or control.

In CORAL, the initial hypertension inclusion criterion of 155 mm Hg was subsequently abandoned during the trial, suggesting that hypertension in many of these patients may have been mild and/or well controlled. Although medical therapy in CORAL was standardized, it also is notable that all patients had their medical therapy adjusted prior to randomization during a roll-in phase to achieve target blood pressure goals of 130/80 in patients with CKD and/or diabetes or 140/90 otherwise. I would suggest that achievement of these blood pressure targets on the study medications (candesartan ± hydrochlorothiazide plus amlodipine-atorvastatin) might be appropriately considered success of medical therapy for patients with hypertension in the absence of renal dysfunction, making it challenging to defend proceeding with revascularization in this scenario.

The study protocol, although well designed from the perspective of attempting to isolate the effect of renal artery angioplasty and stenting, therefore did not uniformly reflect what would be considered responsible utilization of renal revascularization in a real-world environment.

Patient enrollment in CORAL was also very selective; only 947 of the 5,322 patients who were screened went on to be enrolled and randomized. It is likely that at least some of those patients who were not enrolled (especially those who declined to participate or were withdrawn by their physicians) were failing aggressive medical therapy and therefore unwilling to being excluded from angioplasty and stenting through randomization. These limitations aside, however, CORAL does provide some very useful observations that should inform treatment decisions. The results demonstrate the efficacy of contemporary medical therapy for many patients, and show that revascularization offers no additional benefit when medical therapy achieves an acceptable clinical response (defined by stable renal function and reasonable blood pressure control). Additional subgroup analyses of the CORAL data are anticipated, but will likely be underpowered to draw conclusions in the absence of identified revascularization effects.

So when should revascularization be considered for patients with atherosclerotic renal artery stenosis? In general, medical therapy is adequate for most patients and should be implemented prior to any consideration of procedural intervention. Revascularization should be considered only for patients who have failed appropriate, aggressive medical therapy; the medications used in CORAL can certainly be regarded as adequate initial therapy for symptomatic renal artery stenosis, but many providers (including myself) would argue that additional agents should be considered before proceeding with revascularization.

When decline in renal function is the indication for considering revascularization, alternative causes (such as intrinsic renal disease) should diminish enthusiasm for proceeding with angioplasty and stenting, particularly when the anatomic disease distribution does not affect the entire renal mass (as in patients with two kidneys and unilateral stenosis). Appropriate candidates for revascularization include patients with severely impaired renal function (particularly in the setting of a precipitous functional decline) or severe acute blood pressure elevation associated with hypertensive emergency (such as acute congestive heart failure, encephalopathy, acute coronary syndrome, or other signs and symptoms of target organ damage resulting from hypertension and/or volume overload). Continuation of failed medical therapy is often unacceptable to these “no-options” patients as well as their providers, both of whom presumably would be unlikely to accept randomization to ongoing medical management.

Other populations that are not represented within these trials include patients with renal artery restenosis and those with nonatherosclerotic disease; it is therefore important to exercise caution when generalizing these study results to these distinct groups of patients. Enrolling patients with severe symptoms who have failed medical therapy will likely remain challenging for future randomized studies in the absence of alternative treatment options. Although the benefits of renal angioplasty and stenting for these “no-options” patients remain to be proved, the uncertainty of response to revascularization is often easier to accept than the ongoing morbidity and mortality associated with staying the course when medical therapy has failed.

Dr. Matthew A. Corriere is a vascular surgeon at Wake Forest University School of Medicine, Winston-Salem, N.C.

This article developed from a debate held at the 2014 Vascular Annual Meeting.

Percutaneous treatment of renal artery occlusive disease is unnecessary and should be abandoned, except in pediatric cases.

BY GEORGE HAMILTON, M.D.

This position is supported by findings from both the ASTRAL trial (N. Engl. J. Med. 2009;361:1953-62) and the CORAL trial (N. Engl. J. Med. 2013 Nov. 18 [doi:10.1056/NEJMoa1310753]).

The ASTRAL trial, a prospective, randomized comparison of best medical therapy with and without stent angioplasty in more than 800 patients, was the largest trial to date when it began back in the 1990s. The well-known results showed no difference in time to first renal event, first vascular and cardiovascular events, and overall survival. Furthermore, there was no difference in these outcomes among patients with greater than 90% stenosis, with the exception of a possible difference in mortality, which trended toward improvement among those with high-grade stenosis.

We concluded that revascularization in the vast majority of patients is unlikely to improve hypertension control or renal function, and that renal artery stenosis is not pathophysiologically important. We also concluded that there is no point in screening for asymptomatic disease; this was back when every patient was getting screened, and treated primarily on the basis of finding a renal arterial stenosis.

Finally, we concluded that properly applied best medical therapy alone was an extremely good treatment.

Several flaws in the trial garnered extensive criticism, however, and the more rigidly designed CORAL trial was expected to address them. The findings confirmed those of the ASTRAL trial. In more than 900 patients from 88 centers, there was absolutely no benefit of intervention with respect to primary and secondary outcomes, including among those with high-grade stenosis.

We can now see on the basis of extensive level 1 evidence that when added to comprehensive, multifactorial medical therapy, intervention yielded no benefit.

So are there certain patient groups who might benefit more from intervention? Among listed indications are high-grade stenosis (which doesn’t apply any longer); short history of progressive failure (which is quite rare); ACE-induced renal failure (which is also quite rare); difficult-to-control hypertension (there really is no such thing now, except in a tiny percentage of patients); and – the least challenged indication – flash pulmonary edema. These remaining indications move our interventions into a very high risk group of patients.

The current debate is focused almost entirely on endovascular intervention, but a systematic review showed that there is long-term benefit in terms of renal function and hypertension with open procedures. Although overall there is increased mortality, this risk is minimized – and not significantly different from endovascular procedures – in those having only renal revascularization vs. those having concomitant aortic procedures. So open surgery remains a possible treatment option, indeed a recent level 1 study comparing stenting and open surgery, showed better long-term results with open surgery (J. Vasc. Surg. 2009;49:667-75). The authors concluded that surgical reconstruction remains the gold standard in treating renal artery stenosis. Although national data suggest an overall mortality of about 10%, it is much lower at specialist, high-volume centers with mortality rates similar to those of stent angioplasty.

Renal stenting is not a low-risk procedure. In all-comers the complication rates, serious complication rates, and mortality rates are significant with short-term equivalence between focused renal arterial surgery and percutaneous intervention.

Returning to the debate, are either methods of revascularization appropriate? Probably not.

Even in flash pulmonary edema, there is little evidence to support revascularization. Few papers exist suggesting a benefit of revascularization in reduction of flash pulmonary edema, but the patient numbers were small, and there was no benefit in terms of preservation of renal function.

The history of evolution and evaluation of the role of renal revascularization is remarkably similar to that of renal denervation, initially and with considerable conviction thought to be a cure for hypertension. However, when properly assessed by prospective randomized comparison there was found to be absolutely no benefit.

So, given the considerable objective evidence from two major trials and revisiting the basics of the pathophysiology of atherosclerotic renovascular disease, to expect benefit from treating the osteal component of renal artery occlusive disease is at best naive, in my opinion. There remains little clinical evidence of benefit for any indication, with the possible exceptions of ACE-induced renal failure and possibly flash pulmonary edema in the presence of bilateral renal arterial stenoses.

Dr. Hamilton is a professor at the Royal Free London Hospital, University College London, United Kingdom.

There are still indications to treat renal artery occlusive disease

BY MATTHEW A. CORRIERE, M.D.

Although renal artery revascularization has been grossly overutilized and is not indicated in the majority of patients with renal artery stenosis, I perform renal artery revascularization as part of my routine clinical practice and believe that there are many instances where revascularization should be considered, particularly when patients have severe symptoms despite aggressive medical therapy. While neither ASTRAL nor CORAL observed any benefit associated with revascularization, both have important limitations that should be kept in mind when interpreting the results of these trials.

These limitations can be broadly categorized as mismatch between indications for revascularization and clinical endpoints, selection biases favoring enrollment of patients with relatively mild symptoms, and inconsistencies between study protocols and contemporary decision-making strategies.

Given that ASTRAL’s primary outcome was change in renal function (defined by a 20% or greater reduction in the mean slope of the reciprocal of serum creatinine), it is important to remember that the inclusion criteria were renal artery stenosis with unexplained renal dysfunction or poorly controlled hypertension. Patients who had hypertension in the absence of significant renal dysfunction were therefore eligible, and 40% of the randomized participants had preserved baseline renal function (based on a serum creatinine of < 150 micromol/liter). Unlike patients with baseline renal dysfunction (which, in theory, might improve with revascularization), these patients with normal renal function who were treated with revascularization risked decline in renal function resulting from procedure-related adverse events without any real chance of renal function improvement. It would certainly be difficult to justify revascularization for the sake of renal function salvage in these patients, and their inclusion within a randomized trial with change in renal function as its primary outcome is problematic for the same reason.

ASTRAL also had an additional, somewhat unorthodox inclusion criterion: uncertainty on the part of the treating physician that the patient “definitely would have a worthwhile clinical benefit from revascularization.” Exclusion of patients considered likely to benefit from revascularization would seem to ensure a selection bias favoring the null hypothesis; this approach may also explain the large proportion of participants with relatively mild occlusive disease (40% had stenotic lesions that were < 70% in severity).

A high rate of both technical failure (12%) and adverse events (20%) associated with revascularization, asymmetric crossover between treatment groups (86 of the 110 patients who did not receive their randomized intervention were in the revascularization group), and lack of standardized protocol for medical therapy further limit the conclusions that can be drawn from the ASTRAL results.

Although this trial does not provide us with compelling evidence that renal revascularization should be abandoned for patients failing appropriate medical therapy, ASTRAL demonstrated that no benefit should be expected from nonselective use of revascularization, which can be associated with significant rates of both technical failure and major adverse events.

The CORAL trial overcame many of the design limitations for which ASTRAL drew criticism. CORAL’s primary endpoint (freedom from major adverse cardiovascular or renal events) allowed potential benefit for participants with either systolic hypertension or chronic kidney disease as their indication for treatment. Although participants with systolic hypertension as their inclusion criterion had to be on at least two antihypertensive medications, it is important to acknowledge the growing number of indications for these medications related to cardiovascular risk reduction in the setting of diabetes, heart disease, and other diagnoses that may be unrelated to any specific blood pressure target. Number of antihypertensive medications is therefore often a crude and potentially invalid indicator of hypertension severity or control.

In CORAL, the initial hypertension inclusion criterion of 155 mm Hg was subsequently abandoned during the trial, suggesting that hypertension in many of these patients may have been mild and/or well controlled. Although medical therapy in CORAL was standardized, it also is notable that all patients had their medical therapy adjusted prior to randomization during a roll-in phase to achieve target blood pressure goals of 130/80 in patients with CKD and/or diabetes or 140/90 otherwise. I would suggest that achievement of these blood pressure targets on the study medications (candesartan ± hydrochlorothiazide plus amlodipine-atorvastatin) might be appropriately considered success of medical therapy for patients with hypertension in the absence of renal dysfunction, making it challenging to defend proceeding with revascularization in this scenario.

The study protocol, although well designed from the perspective of attempting to isolate the effect of renal artery angioplasty and stenting, therefore did not uniformly reflect what would be considered responsible utilization of renal revascularization in a real-world environment.

Patient enrollment in CORAL was also very selective; only 947 of the 5,322 patients who were screened went on to be enrolled and randomized. It is likely that at least some of those patients who were not enrolled (especially those who declined to participate or were withdrawn by their physicians) were failing aggressive medical therapy and therefore unwilling to being excluded from angioplasty and stenting through randomization. These limitations aside, however, CORAL does provide some very useful observations that should inform treatment decisions. The results demonstrate the efficacy of contemporary medical therapy for many patients, and show that revascularization offers no additional benefit when medical therapy achieves an acceptable clinical response (defined by stable renal function and reasonable blood pressure control). Additional subgroup analyses of the CORAL data are anticipated, but will likely be underpowered to draw conclusions in the absence of identified revascularization effects.

So when should revascularization be considered for patients with atherosclerotic renal artery stenosis? In general, medical therapy is adequate for most patients and should be implemented prior to any consideration of procedural intervention. Revascularization should be considered only for patients who have failed appropriate, aggressive medical therapy; the medications used in CORAL can certainly be regarded as adequate initial therapy for symptomatic renal artery stenosis, but many providers (including myself) would argue that additional agents should be considered before proceeding with revascularization.

When decline in renal function is the indication for considering revascularization, alternative causes (such as intrinsic renal disease) should diminish enthusiasm for proceeding with angioplasty and stenting, particularly when the anatomic disease distribution does not affect the entire renal mass (as in patients with two kidneys and unilateral stenosis). Appropriate candidates for revascularization include patients with severely impaired renal function (particularly in the setting of a precipitous functional decline) or severe acute blood pressure elevation associated with hypertensive emergency (such as acute congestive heart failure, encephalopathy, acute coronary syndrome, or other signs and symptoms of target organ damage resulting from hypertension and/or volume overload). Continuation of failed medical therapy is often unacceptable to these “no-options” patients as well as their providers, both of whom presumably would be unlikely to accept randomization to ongoing medical management.

Other populations that are not represented within these trials include patients with renal artery restenosis and those with nonatherosclerotic disease; it is therefore important to exercise caution when generalizing these study results to these distinct groups of patients. Enrolling patients with severe symptoms who have failed medical therapy will likely remain challenging for future randomized studies in the absence of alternative treatment options. Although the benefits of renal angioplasty and stenting for these “no-options” patients remain to be proved, the uncertainty of response to revascularization is often easier to accept than the ongoing morbidity and mortality associated with staying the course when medical therapy has failed.

Dr. Matthew A. Corriere is a vascular surgeon at Wake Forest University School of Medicine, Winston-Salem, N.C.

This article developed from a debate held at the 2014 Vascular Annual Meeting.

Percutaneous treatment of renal artery occlusive disease is unnecessary and should be abandoned, except in pediatric cases.

BY GEORGE HAMILTON, M.D.

This position is supported by findings from both the ASTRAL trial (N. Engl. J. Med. 2009;361:1953-62) and the CORAL trial (N. Engl. J. Med. 2013 Nov. 18 [doi:10.1056/NEJMoa1310753]).

The ASTRAL trial, a prospective, randomized comparison of best medical therapy with and without stent angioplasty in more than 800 patients, was the largest trial to date when it began back in the 1990s. The well-known results showed no difference in time to first renal event, first vascular and cardiovascular events, and overall survival. Furthermore, there was no difference in these outcomes among patients with greater than 90% stenosis, with the exception of a possible difference in mortality, which trended toward improvement among those with high-grade stenosis.

We concluded that revascularization in the vast majority of patients is unlikely to improve hypertension control or renal function, and that renal artery stenosis is not pathophysiologically important. We also concluded that there is no point in screening for asymptomatic disease; this was back when every patient was getting screened, and treated primarily on the basis of finding a renal arterial stenosis.

Finally, we concluded that properly applied best medical therapy alone was an extremely good treatment.

Several flaws in the trial garnered extensive criticism, however, and the more rigidly designed CORAL trial was expected to address them. The findings confirmed those of the ASTRAL trial. In more than 900 patients from 88 centers, there was absolutely no benefit of intervention with respect to primary and secondary outcomes, including among those with high-grade stenosis.

We can now see on the basis of extensive level 1 evidence that when added to comprehensive, multifactorial medical therapy, intervention yielded no benefit.

So are there certain patient groups who might benefit more from intervention? Among listed indications are high-grade stenosis (which doesn’t apply any longer); short history of progressive failure (which is quite rare); ACE-induced renal failure (which is also quite rare); difficult-to-control hypertension (there really is no such thing now, except in a tiny percentage of patients); and – the least challenged indication – flash pulmonary edema. These remaining indications move our interventions into a very high risk group of patients.

The current debate is focused almost entirely on endovascular intervention, but a systematic review showed that there is long-term benefit in terms of renal function and hypertension with open procedures. Although overall there is increased mortality, this risk is minimized – and not significantly different from endovascular procedures – in those having only renal revascularization vs. those having concomitant aortic procedures. So open surgery remains a possible treatment option, indeed a recent level 1 study comparing stenting and open surgery, showed better long-term results with open surgery (J. Vasc. Surg. 2009;49:667-75). The authors concluded that surgical reconstruction remains the gold standard in treating renal artery stenosis. Although national data suggest an overall mortality of about 10%, it is much lower at specialist, high-volume centers with mortality rates similar to those of stent angioplasty.

Renal stenting is not a low-risk procedure. In all-comers the complication rates, serious complication rates, and mortality rates are significant with short-term equivalence between focused renal arterial surgery and percutaneous intervention.

Returning to the debate, are either methods of revascularization appropriate? Probably not.

Even in flash pulmonary edema, there is little evidence to support revascularization. Few papers exist suggesting a benefit of revascularization in reduction of flash pulmonary edema, but the patient numbers were small, and there was no benefit in terms of preservation of renal function.

The history of evolution and evaluation of the role of renal revascularization is remarkably similar to that of renal denervation, initially and with considerable conviction thought to be a cure for hypertension. However, when properly assessed by prospective randomized comparison there was found to be absolutely no benefit.

So, given the considerable objective evidence from two major trials and revisiting the basics of the pathophysiology of atherosclerotic renovascular disease, to expect benefit from treating the osteal component of renal artery occlusive disease is at best naive, in my opinion. There remains little clinical evidence of benefit for any indication, with the possible exceptions of ACE-induced renal failure and possibly flash pulmonary edema in the presence of bilateral renal arterial stenoses.

Dr. Hamilton is a professor at the Royal Free London Hospital, University College London, United Kingdom.

There are still indications to treat renal artery occlusive disease

BY MATTHEW A. CORRIERE, M.D.

Although renal artery revascularization has been grossly overutilized and is not indicated in the majority of patients with renal artery stenosis, I perform renal artery revascularization as part of my routine clinical practice and believe that there are many instances where revascularization should be considered, particularly when patients have severe symptoms despite aggressive medical therapy. While neither ASTRAL nor CORAL observed any benefit associated with revascularization, both have important limitations that should be kept in mind when interpreting the results of these trials.

These limitations can be broadly categorized as mismatch between indications for revascularization and clinical endpoints, selection biases favoring enrollment of patients with relatively mild symptoms, and inconsistencies between study protocols and contemporary decision-making strategies.

Given that ASTRAL’s primary outcome was change in renal function (defined by a 20% or greater reduction in the mean slope of the reciprocal of serum creatinine), it is important to remember that the inclusion criteria were renal artery stenosis with unexplained renal dysfunction or poorly controlled hypertension. Patients who had hypertension in the absence of significant renal dysfunction were therefore eligible, and 40% of the randomized participants had preserved baseline renal function (based on a serum creatinine of < 150 micromol/liter). Unlike patients with baseline renal dysfunction (which, in theory, might improve with revascularization), these patients with normal renal function who were treated with revascularization risked decline in renal function resulting from procedure-related adverse events without any real chance of renal function improvement. It would certainly be difficult to justify revascularization for the sake of renal function salvage in these patients, and their inclusion within a randomized trial with change in renal function as its primary outcome is problematic for the same reason.

ASTRAL also had an additional, somewhat unorthodox inclusion criterion: uncertainty on the part of the treating physician that the patient “definitely would have a worthwhile clinical benefit from revascularization.” Exclusion of patients considered likely to benefit from revascularization would seem to ensure a selection bias favoring the null hypothesis; this approach may also explain the large proportion of participants with relatively mild occlusive disease (40% had stenotic lesions that were < 70% in severity).

A high rate of both technical failure (12%) and adverse events (20%) associated with revascularization, asymmetric crossover between treatment groups (86 of the 110 patients who did not receive their randomized intervention were in the revascularization group), and lack of standardized protocol for medical therapy further limit the conclusions that can be drawn from the ASTRAL results.

Although this trial does not provide us with compelling evidence that renal revascularization should be abandoned for patients failing appropriate medical therapy, ASTRAL demonstrated that no benefit should be expected from nonselective use of revascularization, which can be associated with significant rates of both technical failure and major adverse events.

The CORAL trial overcame many of the design limitations for which ASTRAL drew criticism. CORAL’s primary endpoint (freedom from major adverse cardiovascular or renal events) allowed potential benefit for participants with either systolic hypertension or chronic kidney disease as their indication for treatment. Although participants with systolic hypertension as their inclusion criterion had to be on at least two antihypertensive medications, it is important to acknowledge the growing number of indications for these medications related to cardiovascular risk reduction in the setting of diabetes, heart disease, and other diagnoses that may be unrelated to any specific blood pressure target. Number of antihypertensive medications is therefore often a crude and potentially invalid indicator of hypertension severity or control.

In CORAL, the initial hypertension inclusion criterion of 155 mm Hg was subsequently abandoned during the trial, suggesting that hypertension in many of these patients may have been mild and/or well controlled. Although medical therapy in CORAL was standardized, it also is notable that all patients had their medical therapy adjusted prior to randomization during a roll-in phase to achieve target blood pressure goals of 130/80 in patients with CKD and/or diabetes or 140/90 otherwise. I would suggest that achievement of these blood pressure targets on the study medications (candesartan ± hydrochlorothiazide plus amlodipine-atorvastatin) might be appropriately considered success of medical therapy for patients with hypertension in the absence of renal dysfunction, making it challenging to defend proceeding with revascularization in this scenario.

The study protocol, although well designed from the perspective of attempting to isolate the effect of renal artery angioplasty and stenting, therefore did not uniformly reflect what would be considered responsible utilization of renal revascularization in a real-world environment.

Patient enrollment in CORAL was also very selective; only 947 of the 5,322 patients who were screened went on to be enrolled and randomized. It is likely that at least some of those patients who were not enrolled (especially those who declined to participate or were withdrawn by their physicians) were failing aggressive medical therapy and therefore unwilling to being excluded from angioplasty and stenting through randomization. These limitations aside, however, CORAL does provide some very useful observations that should inform treatment decisions. The results demonstrate the efficacy of contemporary medical therapy for many patients, and show that revascularization offers no additional benefit when medical therapy achieves an acceptable clinical response (defined by stable renal function and reasonable blood pressure control). Additional subgroup analyses of the CORAL data are anticipated, but will likely be underpowered to draw conclusions in the absence of identified revascularization effects.

So when should revascularization be considered for patients with atherosclerotic renal artery stenosis? In general, medical therapy is adequate for most patients and should be implemented prior to any consideration of procedural intervention. Revascularization should be considered only for patients who have failed appropriate, aggressive medical therapy; the medications used in CORAL can certainly be regarded as adequate initial therapy for symptomatic renal artery stenosis, but many providers (including myself) would argue that additional agents should be considered before proceeding with revascularization.

When decline in renal function is the indication for considering revascularization, alternative causes (such as intrinsic renal disease) should diminish enthusiasm for proceeding with angioplasty and stenting, particularly when the anatomic disease distribution does not affect the entire renal mass (as in patients with two kidneys and unilateral stenosis). Appropriate candidates for revascularization include patients with severely impaired renal function (particularly in the setting of a precipitous functional decline) or severe acute blood pressure elevation associated with hypertensive emergency (such as acute congestive heart failure, encephalopathy, acute coronary syndrome, or other signs and symptoms of target organ damage resulting from hypertension and/or volume overload). Continuation of failed medical therapy is often unacceptable to these “no-options” patients as well as their providers, both of whom presumably would be unlikely to accept randomization to ongoing medical management.

Other populations that are not represented within these trials include patients with renal artery restenosis and those with nonatherosclerotic disease; it is therefore important to exercise caution when generalizing these study results to these distinct groups of patients. Enrolling patients with severe symptoms who have failed medical therapy will likely remain challenging for future randomized studies in the absence of alternative treatment options. Although the benefits of renal angioplasty and stenting for these “no-options” patients remain to be proved, the uncertainty of response to revascularization is often easier to accept than the ongoing morbidity and mortality associated with staying the course when medical therapy has failed.

Dr. Matthew A. Corriere is a vascular surgeon at Wake Forest University School of Medicine, Winston-Salem, N.C.

This article developed from a debate held at the 2014 Vascular Annual Meeting.