Nephrol Dial Transplant (1998) 13: 1690–1695
Nephrology
Dialysis
Transplantation
Original Article
Intravascular ultrasound imaging of atherosclerotic renal arteries:
comparison between in vitro and in vivo studies
Gen Yasuda1, Toshikazu Takizawa1, Izumi Takasaki1, Hiroshi Shionoiri1, Satoshi Umemura1, and
Kiyoshi Shimoyama2
1Second Department of Internal Medicine and 2Department of Pathology, Yokohama City University School of Medicine,
Yokohama, Japan
Introduction
diagnosis of renal artery stenosis (RAS ) resulting in
ischaemic nephropathy and selection of therapeutic
options such as surgical revascularization, percutaneous transluminal angioplasty, or conservative medical
therapy, have largely relied on contrast angiography.
This technique can detect areas of vascular stenosis
and the presence of collateral circulation. However,
angiographic appearance does not give either a precise
description of wall thickness resulting from atherosclerosis or an exact location of thrombi or atheromas on
the lumen surface. Further, contrast angiography
cannot provide a precise diagnosis about the aetiology
of RAS, including atherosclerosis, fibromuscular
dysplasia, or aortitis. These factors require alternative
modalities.
Recently the development of techniques using highfrequency intravascular ultrasound (IVUS) imaging
has allowed us to evaluate vascular lesions. In particular, in the field of ischaemic heart disease, extensive
studies have reported that IVUS provided the potential
to examine not only the architecture of blood vessel
walls but also the composition and location of atherosclerotic plaques in stenotic coronary arteries. On the
other hand, only a few clinical studies [1–3] have used
ultrasound images to assess vascular lesions in the
renal artery. In addition the images of the renal artery
obtained by ultrasound catheters have not yet been
completely evaluated by in vitro studies to confirm
pathological changes.
The aim of this study is twofold: first, to evaluate
the ultrasound images from isolated human renal arteries, and second, to examine how cross-sectional areas
of renal arteries are visualized in hypertensive patients
with atherosclerosis in order to compare the findings
between in vivo and in vitro ultrasound images.
Ischaemic nephropathy leads to secondary hypertension which can resolve with proper treatment. The
Subjects and methods
Abstract
Background. Intravascular ultrasound (IVUS) imaging, a new modality, may be feasible and useful for
the assessment of atherosclerotic renal arteries.
However, comparison between in vivo and in vitro
studies to confirm pathological changes corresponding
with IVUS findings obtained from renal arteries was
not fully evaluated.
Methods. We evaluated ultrasound images of 18 postmortem human renal arteries and cross-sectional IVUS
images of main renal arteries in five patients with renal
artery stenosis (RAS ) or essential hypertension.
Results. In vitro studies have shown that renal-artery
images had three layers when the arteries had fibrous
intimal thickening and medial hypertrophy. Renal
arteries, in which the fibrous intima was not well
developed, showed circumferentially homogenous
bright echoes. In patients with atherosclerotic RAS
and essential hypertension, IVUS images showed
hyperechoic areas in the renal arterial walls, probably
due to atherosclerosis. Typical three-layered ultrasound
appearance was not easily seen during in vivo studies.
Conclusion. Our findings suggest that hyperechoic
images can be a diagnostic clue of atherosclerosis
However, in vitro results do not always correspond
exactly to in vivo findings, and caution is needed when
findings from in vitro IVUS imaging studies are applied
to in vivo studies.
Key words: angiography; angioplasty; atherosclerosis
fibromuscular dysplasia renovascular hypertension
Correspondence and offprint requests to: Gen Yasuda MD, The
Second Department of Internal Medicine, Yokohama City
University, School of Medicine, 3–46 Urafune, Minami-ku,
Yokohama, 232 Japan.
Ultrasound probe
In vitro and in vivo renal artery images were obtained using
a 5-Fr IVUS imaging catheter (Model 54600, Endosonics
© 1998 European Renal Association–European Dialysis and Transplant Association
Intravascular US imaging of renal arteries
corporation, Pleasanton, CA, USA) with a transducer (centre
frequency of 20 MHz) mounted on the tip of a catheter that
provides a 360° scan at 10 f.p.s. Two-dimensional real-time
images of transverse vessels on a video monitor were recorded
on a 1/2-inch videotape and printed by a videographic printer
( UP-850, Sony, Tokyo, Japan).
In vitro studies
Eighteen fresh postmortem renal arteries (from 2.2 to 3.8 cm
in length, mean±SD 2.9±0.6 cm) containing ostial portions
were obtained from nine patients (5 men and 4 women, age
range, 59–74 years, mean±SD 65±7 years). Three patients
had died of liver cancer, two of colon cancer, two of aortic
aneurysm, one of chronic heart failure, and one of multiple
myeloma. No information regarding renal arteries was available from clinical examination. The specimens were separated
at autopsy within 4 h of death and fixed in 10% formaldehyde,
then placed in saline solution at 37°C for the examination.
The ultrasound catheter was placed in the lumen of the renal
arteries to obtain circumferential vascular images along the
entire length of the arteries. After paraffin embedding, the
specimens were transversely sectioned (5-mm thick) at a
portion marked by dye where echo images had been obtained,
and stained with the elastic–Van Gieson’s stain. Specimens
were examined with a microscope (PM-20, Olympus, Tokyo,
Japan) at a 2× magnification to compare histological
findings with the corresponding ultrasound images.
In vivo studies
Patient selection. Three patients with RAS and two patients
with essential hypertension participated in this study. They
were hospitalized and were given a constant diet containing
120 mmol/day of NaCl. The diagnosis of RAS and essential
hypertension was made based on the criteria we have reported
previously [4]. All patients had normal to slightly impaired
renal function. Informed consent was obtained from each
patient. A brief background for each patient is set out below.
Case 1. A 68-year-old man had been treated for essential
hypertension for about 5 years. His blood pressure had been
controlled at approximately 150/100 mmHg. Blood sample
examinations revealed low serum potassium levels and a
slightly elevated plasma renin activity, suggesting that he
had renovascular hypertension. For close medical supervision, he was admitted to our hospital. Renal angiography
revealed that the left renal artery had a 60% diameter stenosis
that was calculated by dividing the diameter at the stenosis
with that at angiographically normal distal portions.
Case 2. A 73-year-old woman had been treated for severe
systolic hypertension, angina, and intermittent claudication
due to arteriosclerosis obliterans. She also had hyperlipidaemia and hypercholesterolaemia. A combination of diuretics,
calcium antagonists, and angiotensin-converting enzyme
inhibitors did not reduce her blood pressure sufficiently. RAS
resulting from atherosclerosis was suspected on the basis of
high plasma renin activity and mild ipsilateral renal dysfunction that was confirmed by intravascular pyelography.
Angiography disclosed advanced systemic atherosclerosis
including bilateral ostial RAS (>95%).
Case 3. An 18-year-old male student was incidentally noted
to have juvenile-type hypertension when he was admitted for
treatment of hepatitis C. His family history was not significant. Clinical and laboratory examinations revealed that he
had hyperreninaemia, hyperaldosteronaemia and hypokalaemia, suggesting that his hypertension was caused by RAS
1691
due to fibromuscular dysplasia. Angiography exhibited 99%
reduction in diameter in the middle of the left main renal
artery.
Case 4. A 62-year-old man had hypertension with hyperreninaemia and hyperaldosteronaemia. Renal scintigram
showed a discrepancy in kidney size, suggesting RAS.
However, renal angiography revealed no remarkable vascular
stenosis. Eventually he was diagnosed as having essential
hypertension.
Case 5. A 70-year-old man was admitted to our hospital
to evaluate hypertension that had persisted for more than 18
years. An abdominal vascular murmur and increased plasma
renin activity suggested that he had RAS. To establish a
diagnosis, he underwent selective renal vein blood sampling
for plasma renin activity and renal angiography. These
examinations revealed that he did not have RAS but did
have essential hypertension.
Procedure
Renal angiography was performed using a standard method
through a femoral artery approach. All procedures were
performed under fluoroscopic guidance. Renal artery stenosis
was determined by intra-arterial digital subtraction renal
angiography. An 8-Fr guiding catheter (RESS, Schneider,
Minneapolis, MN, USA) was placed with the tip at the
orifice of the renal artery. The IVUS catheter that was
advanced antegrade or retrograde over a 0.014-inch guidewire
was introduced through the guiding catheter. The guidewire
was placed peripherally in the main renal artery and was
used to control the ultrasound catheter. The IVUS catheter
was attached to the IVUS unit to obtain real-time crosssectional images. Ultrasound images of the entire length of
renal arteries were displayed on the video monitor as the
catheter was moved with the aid of the guidewire. Renal
artery stenosis greater than 75% in diameter was considered
an indication for percutaneous transluminal angioplasty.
Angioplasty was performed by inflating balloons with a
diameter equal to the angiographic lumen size of the artery
next to the stenosis. Balloons were repeatedly inflated for
30–60 s using a manual inflation system (Baxter Healthcare
Corp., Santa Ana, CA, USA). Inflation pressures were
increased in a stepwise fashion to 10 atm.
Results
In vitro studies
The renal artery specimens gave a circumferential
vascular echo appearance around catheter ring images
and an echo-free zone, which corresponded to the
vascular lumen. A representative ultrasound image is
shown in Figure 1a. Outlines of transverse sections
were consistent with echo appearances ( Figure 1b). A
circumferential hypoecho reflectance corresponded to
the muscular media and existed between inner and
outer bright echo zones. These two bright echo layers
seen in the internal surface and external portion corresponded to the intima and adventitia respectively. The
three-layer ultrasound appearance of the vessel was
more prominent when the fibrous thickening of the
intima was present. A thin intima had low echogenicity
( Figure 1c), while a fibrous intima had an increased
1692
G. Yasuda et al.
echo density, showing a partial bright echo reflectance
( Figure 1d) or the three-layer structure ( Figure 1e).
In vivo studies
(a)
(b)
(c)
The ultrasound catheter manipulation was carried out
without major complications, and ultrasound images
were obtained with the catheter advanced either
antegrade or retrograde. The catheter was not easily
positioned coaxial to the long axis of the vessel lumen,
but rather tended to be placed eccentrically within the
vascular lumen when the probe was in the proximal
site of the renal arteries, probably due to the relatively
large diameter of the vascular lumen. With the support
of a guidewire it became easier to keep the catheter
central and parallel to the renal arteries at the distal
and relatively straight portions. Eccentric position
induced an oval-shaped echo reflectance, showing
bright echoes in the closer walls. It was difficult to
obtain desirable echo images within tortuous parts of
arteries even with the aid of a guidewire.
Case 1 was a patient with a 60% renal artery stenosis.
In this patient the ultrasound catheter was introduced
beyond the stenotic portion and was moved back to
the proximal site of the lesion. At each point, crosssectional vascular images were obtained (Figure 2). A
three-layer appearance was not seen over the entire
length of the renal artery. In case 2 ultrasound images
were not obtained because almost complete ostial
stenosis resulting from atherosclerosis made it impossible to pass the ultrasound probe into the renal arteries.
The patient in case 3 had RAS caused by fibromuscular
dysplasia that was examined as a comparison to RAS
due to atherosclerosis. In this patient, we did not
attempt to obtain ultrasound images prior to angioplasty for fear the catheter would completely obstruct the
stenotic vascular lumen. Angioplasty was performed
by inflating balloons three times for 60 s at 10 atm.
and dilated the stenosis by 30% in diameter. The
ultrasound images after balloon therapy near the site
of dilatation showed a bright and uniform echo
reflectance, a finding suspected to occur in fibromuscular dysplasia (Figure 3). To generate an optimal
(d)
(e)
Fig. 1a–e. In vitro renal vascular echo images and corresponding
pathological findings (elastic-Van Gieson’s stain, original magnification, ×2); calibrator indicates a 1-mm scale. (a) Representative
ultrasound image with the circumferential three-layered appearance.
Intima is seen between lumen–tissue interface and low-echo layer of
media. The media is the hypoechoic zone between the bright echoes
of the intima and adventitia (71-year-old man, liver cancer). (b)
Artificially deformed artery. The outline of the vascular specimen
corresponds with the echogenic vascular appearance (59-year-old
man, liver cancer). (c) As the intima is not hypertrophied, the
intima–media interface is obscure (61-year-old woman, chronic heart
failure). (d ) Increase in wall thickness between the 12 and 3 o’clock
positions seen in the histological section is characterized by high
echogenicity in the IVUS image. The partially separated intima from
media is an artefact (74-year-old man, multiple myeloma). (e)
Boundaries between well-thickened intimal and medial layers and
between medial and external layers are seen circumferentially, showing the three-layered appearance (64-year-old man, aneurysm of the
abdominal aorta).
Intravascular US imaging of renal arteries
1693
Fig. 2. In vivo renal vascular echo images and corresponding vascular portions in a 68-year-old man with renal artery stenosis. Arrows
(1,2,3) denote each portion, where a radiopaque marker of the intravascular ultrasound (IVUS) catheter can be observed, corresponding
to the ultrasound images.
Fig. 3. A young patient (18-year-old man) with renal artery stenosis
due to fibromuscular dysplasia. The ultrasound image (right panel )
was obtained after percutaneous transluminal renal angioplasty,
showing a uniform echo reflectance.
circular echo appearance, the ultrasound gain settings
were increased. In cases 4 and 5, partial three-layer
appearances were seen. The closer portion of the intima
tended to show bright echo reflectance (Figure 4).
Discussion
According to earlier in vitro IVUS studies [5,6 ], the
typical image of arteries gave a three-layered appearance, showing a thin inner echo-dense layer, a thick
echo-lucent middle layer, and a thin outer brightecho layer, corresponding to an intima, media, and
adventitia respectively. However, not all arteries have
been recognized to have characteristic three layers.
Whether the distinct three-layered appearance is seen
Fig. 4. A 62-year-old man with essential hypertension. A partial
three-layered appearance is seen from the 12 to 2 o’clock positions.
The closer portion of the intima tends to show a bright echo
reflectance.
depends on the anatomy of the vascular structure [7]
and the quality of the IVUS catheter. Nishimura et al.
[8] have reported that renal arteries anatomically
belong to the type of musculoelastic arteries which
give a homogeneous ultrasound appearance. In the
present examination of dissected renal arteries, the
images of most arteries showed three layers when those
arteries had fibrous intimal thickening and medial
hypertrophy. The remaining renal arteries, in which
intimal thickening was not well developed, revealed
circumferentially homogeneous bright echoes. Similar
findings were also reported in coronary arteries [9,10],
1694
in which ultrasound images were related to disease and
age. Thus these findings indicate that sclerotic vessels
are apt to demonstrate greater echo-reflectance and
the characteristic three-layer reflection.
During in vivo studies, it was easier to obtain vascular
echoes in elderly patients than in younger patients
because their arteries were sclerotic, giving brighter
echoes. Case 1 was a 68-year-old hypertensive patient
with 60% stenosis of the renal artery due to atherosclerosis. In this patient, the echo appearances exhibited
hyperechoic areas, probably due to atherosclerosis or
calcium deposits. Meanwhile case 3 was a young
patient with fibromuscular dysplasia, who showed a
homogeneous but relatively lower echo reflectance
along the vessel wall. This suggests that the difference
between echo reflectance can be a diagnostic clue to
distinguish atherosclerosis from fibromuscular dysplasia. Although sweeping conclusions cannot be drawn
from one case of fibromuscular dysplasia, the observation is tantalizing and it should be established in larger
studies as to whether the difference between echo
reflectance is diagnostic in distinguishing atherosclerosis from fibromuscular dysplasia.
When patients had severe RAS or total occlusion, it
was technically hard to get circumferential IVUS
images. We failed to observe the typical three-layer
appearance in these cases. Meanwhile, Sheikh et al. [1]
observed the three-layered IVUS images in normal
renal arteries. The degree of pathological changes in
vascular walls and differences in the imaging systems
employed may have caused the variability in findings.
It is essential to confirm pathological changes of
renal arteries corresponding to IVUS images observed
during in vivo studies. The current studies suggested
that the findings obtained in the in vitro studies did
not completely correspond with those observed in the
in vivo studies. The technical limitations of in vivo
catheter manipulation must be the reason why whole
layers of renal arteries are difficult to visualize. Further,
considerable differences in circumstances between the
in vivo and in vitro studies may produce the differing
IVUS images. Saline solution that was used instead of
circulating blood in the in vitro studies may yield
different echo reflectance. We examined the specimens
after formaldehyde fixation to avoid artefacts induced
by tissue shrinkage due to formaldehyde. It is controversial whether formaldehyde fixation affects echo
reflectivity. Potkin et al. [9] reported that the influence
of formaldehyde fixation on ultrasound images was
quantitatively and qualitatively minor. On the other
hand, Gussenhoven et al. [7] noted an increase in echo
strength produced by formaldehyde. Other possibilities
included, in the in vivo studies, pulsatile vascular
movement prevents fine stop frame images and the
transducer may not be fixed in the blood stream;
perivascular tissues and organs also influence echogenicity; and high pressure in the vascular lumen may
affect ultrasound images [10]. Overall, our results
suggest that caution is needed to apply information
obtained from in vitro echo imaging studies to in
vivo studies.
G. Yasuda et al.
The incidence of sclerotic RAS among elderly
patients with hypertension is high [11], and consequently the incidence of chronic renal failure has
been rising. The success rate of renal angioplasty in
these elderly patients with severe renal impairment has
been low [12]. However, with the improvement in
devices and techniques used for angioplasty, this
therapy is becoming a viable option for elderly patients.
Nevertheless, catheter-based interventional therapy for
patients with atherosclerotic RAS is plagued by higher
occurrence rates of complications [13] including dissection, thrombosis, and haemorrhage. The restenosis rate
also remains higher in patients with atherosclerotic
RAS than in patients with fibromuscular disease [14].
With contrast angiography, one cannot identify the
vascular damage produced by the interventional procedure. To identify the vascular lesions following the
intervention, IVUS yields more precise information on
the structure of renal arteries. Furthermore, subsequent
adequate long-term follow-up treatment based on echo
appearance, such as anticoagulant therapy, can be
determined.
In the present study we reported the echo appearance
of dissected renal arteries and the IVUS images
obtained from each patient with renal artery atherosclerosis or fibromuscular dysplasia, which are the two
major causes of renovascular hypertension. We failed
in the IVUS examination in two of five patients. When
arteries have severe stenosis,it is difficult and risky to
examine IVUS findings by placing catheters before
interventional therapy. On the other hand, in such
cases it must be useful to examine IVUS findings after
angioplasty. However, the opportunities to confirm the
pathological changes in renal arterial walls corresponding to the in vivo IVUS findings obtained from the
same arteries are rare. Further information is required
for IVUS imaging before we can conclude whether it
is a feasible and beneficial modality for patients with
RAS.
Acknowledgement. We thank Dr S. Hayashi for his suggestion in
evaluating the IVUS imaging.
References
1. Sheikh KH, Davidson CJ, Newman GE, Kisslo KB, Schwab
SJ. Intravascular ultrasound assessment of the renal artery. Ann
Intern Med 1991; 115: 22–25
2. Isner JM, Kaufman J, Rosenfield K et al. Combined physiologic
and anatomic assessment of percutaneous revascularization
using a Doppler guidewire and ultrasound catheter. Am J Cardiol
1993; 71: 70D–86D
3. Elkayam U, Cohen G, Gogia H et al. Renal vasodilatory effect
of endothelial stimulation in patients with chronic congestive
heart failure. J Am Coll Cardiol 1996; 28: 176–182
4. Yasuda G, Nagasawa T, Umemura S, Shionoiri H, Ishii M. The
impaired control of plasma renin activity in hypertensive patients
with end-stage renal disease due to chronic glomerulonephritis.
Clin Nephrol 1994; 42: 300–308
5. Pandian NG. Intravascular and intracardiac ultrasound imaging.
Circulation 1989; 80: 1091–1094
6. Mallery JA, Tobis JM, Griffith J et al. Assessment of normal
and atherosclerotic arterial wall thickness with an intravascular
ultrasound imaging catheter. Am Heart J 1990; 119: 1392–1400
Intravascular US imaging of renal arteries
7. Gussenhoven EJ, Essed CE, Lancée CT et al. Arterial wall
characteristics determined by intravascular ultrasound imaging:
An in vitro study. J Am Coll Cardiol 1989; 14: 947–952
8. Nishimura RA, Edwards WD, Warnes CA et al. Intravascular
ultrasound imaging: In vitro validation and pathologic correlation. J Am Coll Cardiol 1990; 16: 145–154
9. Potkin BN, Bartorelli AL, Gessert JM et al. Coronary artery
imaging with intravascular high-frequency ultrasound.
Circulation 1990; 81: 1575–1585
10. Fitzgerald PJ, Goar EG, Connolly AJ et al. Intravascular
ultrasound imaging of coronary arteries. Circulation 1992; 86:
154–158
1695
11. Holley KE, Hunt JC, Brown AL, Kincaid OW, Sheps SG. Renal
artery stenosis. Am J Med 1964; 37: 14–22
12. Scoble JE, Maher ER, Hamilton G, Dick R, Sweny P, Moorhead
JF. Atherosclerotic renovascular disease causing renal impairment – a case for treatment. Clin Nephrol 1989; 31: 119–122
13. Sos TA, Pickering TG, Sniderman K et al. Percutaneous transluminal renal angioplasty in renovascular hypertension due to
atheroma or fibromuscular dysplasia. N Engl J Med 1983;
309: 274–279
14. Jensen G, Zachrisson B, Delin K, Volkmann R, Aurell M.
Treatment of renovascular hypertension: One year results of
renal angioplasty. Kidney Int 1995; 48: 1936–1945
Received for publication: 13.11.97
Accepted in revised form: 11.3.98