NDT Advance Access originally published online on December 15, 2006
Nephrology Dialysis Transplantation 2007 22(3):689-692; doi:10.1093/ndt/gfl673
© The Author [2006]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
Poor performance of diagnostic tests for atherosclerotic renal artery stenosis—discrepancies between stenosis and renal function
Thomas Klaas Alphons Wierema,
Abraham Anthonie Kroon and
Peter Wilhelmus de Leeuw
Department of Medicine, University Hospital Maastricht and Cardiovascular Research Institute Maastricht (CARIM), Maastricht, The Netherlands
Correspondence and offprint requests to: Peter W. de Leeuw, MD, PhD, FAHA, Department of Medicine, University Hospital Maastricht, PO Box 5800, 6202 AZ Maastricht, The Netherlands. Email: p.deleeuw{at}intmed.unimaas.nl
Keywords: hypertension; renal artery stenosis; screening
 |
Introduction
|
|---|
Both from a pathophysiological point of view and from a clinical
standpoint, atherosclerotic renal artery stenosis (RAS) remains
an enigmatic disorder. With its capricious clinical presentations,
it is often overlooked and diagnosed only at a very late stage.
To a certain degree, the latter may be attributable to a lack
of enthusiasm amongst clinicians to search for RAS. The basis
for this nihilistic attitude mainly lies in the failure of several
prospective studies to show a large enough benefit of surgery
or angioplasty (with or without stent placement) on blood pressure
and renal function. Surely, exposing a patient to diagnostic
tests can be justified only when the outcome of these tests
are relevant for further clinical decision making. In the case
of RAS, however, the situation is a bit more complicated than
outcome trials would have us believe. Indeed, even though the
effect of balloon angioplasty may be relatively small [
1], there
are probably subgroups of patients who would benefit more than
others from such treatment [
2]. Moreover, at present, we do
not know whether the seemingly disappointing results of angioplasty
are attributable to an inherent ineffectiveness of this form
of treatment, or to inappropriate indications. Since the presence
of atherosclerotic RAS is associated with an excessive cardiovascular
risk [
3], it is also possible that even if the treatment did
improve blood pressure control and renal function, those effects
would still come too late. Given that revascularization is usually
limited to patients with more than 50%, or sometimes even 70%
stenosis, it may well be that by the time this degree of stenosis
has been reached, ischaemia-induced intrarenal lesions would
have become irreversible. A critical appraisal of the current
state of affairs is therefore mandatory. In this respect, we
should focus on some pathophysiological aspects of RAS that
may underlie the fallibility of diagnostic modalities and the
poor response to treatment.
|
When should we think of renal artery stenosis?
|
|---|
The two main causes of RAS are atherosclerosis and fibromuscular
dysplasia (FMD). Both conditions clearly differ with respect
to patient characteristics and prognosis. For instance, FMD
is considered to be a disease of the young, especially women,
with a good chance of recovery after percutaneous dilatation.
The latter is supported by data from Alhadad and coworkers [
4]
who, in a retrospective analysis of 69 patients (mean age 44
years), found that nearly 25% of patients were completely cured,
while the remainder showed significant falls in blood pressure
and serum creatinine and a reduced need for antihypertensive
drugs. Whether the concept of FMD occurring mainly in young
female patients will hold is debatable, especially since we
encounter an increasing number of older and male patients with
FMD [
5]. Surely if one looks for FMD only in young females,
the association becomes a self-fulfilling prophecy which tends
to deny that older women, who are first diagnosed with hypertension
at a later age, may have had the disorder all their lives. So
far, no clues have been identified that are sensitive and specific
enough to alert the physician to the diagnosis of FMD.
Atherosclerotic RAS, on the other hand, may be suspected in elderly patients with hypertension and renal dysfunction as key symptoms. Severe or recent onset hypertension, abdominal bruit, male gender, flash pulmonary oedema, hypercholesterolaemia, loss of renal function after treatment with an Angiotensin Converting Enzyme (ACE) inhibitor or an angiotensin receptor antagonist, a history of tobacco use and atherosclerosis elsewhere in the body are clinical clues which may arouse a suspicion of RAS. Yet, in autopsy studies, a significant number of renal artery lesions has been reported, which were apparently not suspected on clinical grounds and which were not associated with hypertension during life [6]. In a general population, older than 65 years and without renal disease, a prevalence of 6.8% was noted [7]. Clinical prediction rules such as the one proposed [8] and recently validated [9] by Krijnen et al. [8,9] may assist the physician in the selection of patients who need to be investigated further, but even then the chance of a positive result is not much greater than 30%. On the other hand, the incidence of finding a lesion in the renal arteries could be as high as 46%, when patients with coronary artery disease are screened [10]. So, one should still be wary of rejecting RAS in a patient who does not meet any of the aforementioned criteria.
|
What do screening tests teach us?
|
|---|
If clinical symptomatology is so fallible, one is inclined to
put greater emphasis on
screening tests. In essence, one can
perform either an anatomical (imaging) test or a functional
test. However, none of the tests which have been developed in
the past 30 years has proved to be reliable enough. As far as
imaging is concerned, only one study has directly compared modern
techniques (CT angiography and MR angiography) with the gold
standard of intra-arterial digital subtraction angiography in
a sufficiently large population of hypertensive patients [
5].
This multicentre trial of the RADISH-group clearly showed that
the current non-invasive imaging techniques are not sensitive
or specific enough to pick up RAS, although the sensitivity
of these tests may improve in selected groups of patients with
a high a priori chance of having the abnormality. Nonetheless,
imaging techniques provide better results than functional tests
[
11]. Conventional wisdom makes us believe that the lack of
accuracy of these tests is due to technical imperfections and/or
poor quality of test characteristics, because we are not inclined
to question the validity of the pathophysiological concepts
upon which the tests are based. However, if one accepts that
our current understanding of RAS and its sequelae is incomplete,
it seems equally possible that the test results should be interpreted
in another way. For instance, renography with or without ACE-inhibition,
has proven too unreliable for the diagnosis of (functional)
RAS, even though the rationale of the test seems sound. Rather
than dismissing the test as being inaccurate, we may have to
conclude that the test result represents a mechanistic phenomenon
that we do not yet fully understand. For instance, the captopril
test is likely to unravel the degree of renin activation, but
to link the result to the presence or absence of RAS is just
one step too far. We have to rethink, therefore, what exactly
the test is measuring. In this regard, it is essential that
we pay closer attention to the left–right differences
in kidney function.
|
Our faulty concept of symmetry
|
|---|
In the era of the Renaissance, symmetry was the trademark of
the successful artist. Even the human body was once perceived
as a construction of perfect symmetry. Modern science, however,
has taught us that there are subtle, yet demonstrable differences
between the left and the right side of the body. Although we
have come to recognize these differences, we keep adhering to
a physiological model in which twin organs such as the kidneys
have similar functions. As a corollary, we consider the results
of screening tests for (unilateral) RAS only abnormal when there
is an asymmetric outcome. In all likelihood, this concept is
wrong. Van Onna and coworkers [
12], for instance, have demonstrated
that in patients with essential hypertension and no angiographic
abnormalities, significant differences exist in renal blood
flow between both kidneys and that in the majority of cases,
left renal flow is lower than right renal flow. Perhaps, if
we were able to measure renal perfusion continuously, we would
see that in the long run the differences would cancel out, but
in the absence of such data we cannot accept a priori that kidney
function is symmetrical. Stretching this reasoning a little
further, it is clear that an asymmetric result of a functional
test does not necessarily point towards the presence of RAS.
Conversely, we cannot exclude the possibility that there are
patients with RAS who have completely symmetric renal perfusion.
Recent (as yet unpublished) data from our group, indeed, substantiate
this view. Also, endothelial function of renal vessels may differ
depending upon whether stenosic lesions are unilateral or bilateral
[
13]. Although we do not know the mechanisms which cause the
left–right difference in renal function, the phenomenon
per se may already offer some explanation for the poor performance
of diagnostic tests in RAS.
|
The fallacy of clinically significant stenosis
|
|---|
The concept of clinically significant stenosis is based upon
old observations in experimental animals, showing that the renal
artery needs to be occluded by
80% before renal blood flow drops
or systemic pressure rises [
14]. Although haemodynamically interesting,
these experiments in dogs hardly mirror the human situation,
in which the atherosclerotic process gradually obliterates the
lumen over a long period of time. The same process probably,
also affects smaller vessels downstream from the main artery
[
15]. Therefore, the current practice to consider a stenosis
as clinically relevant only when the lumen is narrowed by 50%
or, more frequently 70%, lacks sufficient clinical evidence.
On the contrary, we know that atherosclerotic RAS is a progressive
disease [
16] which may ultimately lead to renal parenchymal
atrophy [
17]. Such a progressive disease would suggest that
renal functional characteristics worsen gradually over time.
So far however, relationships between the degree of stenosis
and renal functional measurements have not been found [
15].
If such relationships were present, they could never be strong,
for a number of reasons. First, the estimate of a three-dimensional
luminal obstruction on a two-dimensional plain film is inaccurate,
because atherosclerotic lesions are seldom fully circular or
regular. Second, there is a significant inter- and intra-observer
variability in the estimation of stenosis [
18]. Third, renal
functional studies such as renal blood flow, renin production
and creatinine extraction are complicated and one would need
a large group of patients in whom all such variables are measured
under standardized conditions to demonstrate a relationship,
if any exists. With these prerequisites, our group has demonstrated
that there is indeed a weak but significant positive correlation
between the degree of stenosis and renin production and a negative
one with renal blood flow and serum creatinine. Besides these
theoretical considerations, there are other data which suggest
that RAS obliterating <50% of the lumen may already have
clinical significance [
19,
20]. Furthermore, pressure gradients
over a stenosis have been shown to correlate with the severity
of stenosis and functional parameters also in low-grade stenosis
[
21]. Therefore, viewed from different angles, there are enough
indications that renal function may already be at stake in what
is considered non-significant RAS. In other words, the term
‘clinically significant stenosis’ should be abandoned
until we understand more about the relationships between an
impaired renal blood supply and disturbances of kidney function
or cardiovascular regulating systems.
|
Where should we go from here?
|
|---|
Putting all the pieces together, we are beginning to see that
there is no unequivocal link between the presence of anatomic
lesions in the renal artery and its functional sequelae. What
we can conclude, though, is that the traditional concept of
a critical stenosis leading to renovascular hypertension is
not tenable. Also, the implicit assumption of symmetrically
functioning kidneys needs revision. If one is ready to accept
that, it becomes less difficult to explain why classical screening
tests for RAS fail to predict the presence of this abnormality
with enough precision. However, the results of these screening
tests may turn for the better if they reveal the magnitude of
certain pathophysiological mechanisms, which are relevant for
maintaining optimal perfusion and filtration in the face of
an obstructed flow. If it is true, as some studies suggest,
that even low-grade stenosis is associated with alterations
in renal function, however small, it is imperative to direct
more research towards the early stages of RAS. Given the rather
poor results of (late) revascularization [
22], we may even have
to face the possibility that we should intervene at an earlier
stage to preserve kidney function and to prevent (or treat)
renovascular hypertension. However, before we can recommend
angioplasty in patients with low-grade stenosis, we need a large
randomized trial to investigate whether it is beneficial and
cost-effective to do so. As long as the results of such a trial
are not available, we cannot do much more than rely on our clinical
acumen, and select for angioplasty only those patients with
RAS in whom hypertension is resistant to treatment or who respond
to a lowering of blood pressure with a decline in kidney function.
Other than a treatment trial, we also need to refine our methods
to pick up subtle changes in renal perfusion, and to find ways
to distinguish disease confined to the main renal artery from
that also involving the intrarenal vasculature. However, whether
it is worth diagnosing RAS at all depends entirely upon the
results of trials such as the one suggested above, or major
outcome trials like the CORAL study [
22]. Because knowledge
cannot advance without well-designed trials in patients with
RAS, everybody dealing with this condition should be encouraged
to participate in such trials.
Conflict of interest statement. None declared.
|
References
|
|---|
- van Jaarsveld BC, Krijnen P, Pieterman H, et al. (2000) The effect of balloon angioplasty on hypertension in atherosclerotic renal-artery stenosis. Dutch Renal Artery Stenosis Intervention Cooperative Study Group. N Engl J Med 342:1007–1014.[Abstract/Free Full Text]
- Garovic VD and Textor SC. (2005) Renovascular hypertension and ischemic nephropathy. Circulation 112:1362–1374.
- Uzu T, Takeji M, Yamada N, et al. (2002) Prevalence and outcome of renal artery stenosis in atherosclerotic patients with renal dysfunction. Hypertens Res 25:537–542.[CrossRef][ISI][Medline]
- Alhadad A, Mattiasson I, Ivancev K, Gottsater A, Lindblad B. (2005) Revascularisation of renal artery stenosis caused by fibromuscular dysplasia: effects on blood pressure during 7-year follow-up are influenced by duration of hypertension and branch artery stenosis. J Hum Hypertens 19:761–767.[CrossRef][ISI][Medline]
- Vasbinder GB, Nelemans PJ, Kessels AG, et al. (2004) Accuracy of computed tomographic angiography and magnetic resonance angiography for diagnosing renal artery stenosis. Ann Intern Med 141:674–682 discussion 682.[Abstract/Free Full Text]
- Kuroda S, Nishida N, Uzu T, et al. (2000) Prevalence of renal artery stenosis in autopsy patients with stroke. Stroke 31:61–65.[Abstract/Free Full Text]
- Hansen KJ, Edwards MS, Craven TE, et al. (2002) Prevalence of renovascular disease in the elderly: a population-based study. J Vasc Surg 36:443–451.[CrossRef][ISI][Medline]
- Krijnen P, Van Jaarsveld BC, Steyerberg EW, Man in’t Veld AJM, Schalekamp MADH, Habbema JDF. (1998) A clinical prediction rule for renal artery stenosis. Ann Intern Med 129:705–711.[Abstract/Free Full Text]
- Krijnen P, Steyerberg EW, Postma CT, Flobbe K, de Leeuw PW, Hunink MG. (2005) Validation of a prediction rule for renal artery stenosis. J Hypertens 23:1583–1588.[ISI][Medline]
- Rihal CS, Textor SC, Breen JF, et al. (2002) Incidental renal artery stenosis among a prospective cohort of hypertensive patients undergoing coronary angiography. Mayo Clin Proc 77:309–316.[ISI][Medline]
- Vasbinder GBC, Nelemans PJ, Kessels AGH, Kroon AA, De Leeuw PW, Van Engelshoven JMA. (2001) Diagnostic tests for renal artery stenosis in patients suspected of having renovascular hypertension: a meta-analysis. Ann Intern Med 135:401–411.[Abstract/Free Full Text]
- Van Onna M, Houben AJ, Kroon AA, et al. (2003) Asymmetry of renal blood flow in patients with moderate to severe hypertension. Hypertension 41:108–113.[Abstract/Free Full Text]
- Wierema TK, Houben AJ, Kroon AA, et al. (2001) Nitric oxide dependence of renal blood flow in patients with renal artery stenosis. J Am Soc Nephrol 12:1836–1843.[Abstract/Free Full Text]
- Finke R, Gross R, Hackenthal E, Huber J, Kirchheim HR. (1983) Threshold pressure for the pressure-dependent renin release in the autoregulating kidney of conscious dogs. Pflugers Arch 399:102–110.[CrossRef][ISI][Medline]
- Wright JR, Shurrab AE, Cheung C, et al. (2002) A prospective study of the determinants of renal functional outcome and mortality in atherosclerotic renovascular disease. Am J Kidney Dis 39:1153–1161.[CrossRef][ISI][Medline]
- Caps MT, Perissinotto C, Zierler RE, et al. (1998) Prospective study of atherosclerotic disease progression in the renal artery. Circulation 98:2866–2872.
- Caps MT, Zierler RE, Polissar NL, et al. (1998) Risk of atrophy in kidneys with atherosclerotic renal artery stenosis. Kidney Int 53:735–742.[CrossRef][ISI][Medline]
- Schreij G, de Haan MW, Oei TK, Koster D, de Leeuw PW. (1999) Interpretation of renal angiography by radiologists. J Hypertens 17:1737–1741.[CrossRef][ISI][Medline]
- Schoenberg SO, Bock M, Kallinowski F, Just A. (2000) Correlation of hemodynamic impact and morphologic degree of renal artery stenosis in a canine model. J Am Soc Nephrol 11:2190–2198.[Abstract/Free Full Text]
- Schreij G, Ritsema GH, Vreugdenhil G, de Leeuw PW. (1996) Stenosis and renographic characteristics in renovascular disease. J Nucl Med 37:594–597.[Abstract/Free Full Text]
- Gross CM, Kramer J, Weingartner O, et al. (2001) Determination of renal arterial stenosis severity: comparison of pressure gradient and vessel diameter. Radiology 220:751–756.[Abstract/Free Full Text]
- Dworkin LD. (2006) Controversial treatment of atherosclerotic renal vascular disease: the cardiovascular outcomes in renal atherosclerotic lesions trial. Hypertension 48:350–356.[Free Full Text]
Received for publication: 7. 2.06
Accepted in revised form: 18.10.06
CiteULike 
Connotea 
Del.icio.us What's this?
Top
Introduction
When should we think...