Eur. Radiol. (2001) 11: 1902±1915 DOI 10.1007/s003300101012

Alexander V. Zubarev

Published online: 30 August 2001  Springer-Verlag 2001 * Categorical Course ECR 2002

)

A. V. Zubarev ( ) Department of Radiology, Postgraduate Education and Research Center, Government Medical Center, M. Timoshenko 21, 121359 Moscow, Russia Phone: +7-95-7 64 23 92 Fax: +7-95-1 49 58 27

U LT R A S O U N D *

Ultrasound of renal vessels

Abstract Kidneys are known as well-perfused organs and may undergo a variety amount of vascular pathological conditions such as renal artery stenosis, renal vein thrombosis, arteriovenous fistula, and aneurysms. Sonography is usually the first imaging method for renal vascular diseases. Modern US machines are now able to outline with great detail both main renal vessels and intraparenchymal vasculature of the kidney using color and power Doppler techniques.

Introduction Sonography of the kidneys is well known as an excellent modality and initial procedure in the examination of patients with renal pathology. Although able to provide a vast array of information about the morphology of the kidney and the renal sinus, conventional US is not generally helpful in providing evaluation of changes affecting the renal vessels. Such information can be obtained through Doppler techniques. Modern commercially available machines are now able to outline with great detail both main renal vessels and intraparenchymal vasculature of the kidney using color and power Doppler techniques. By means of spectral analysis, Doppler US can provide evaluation of flow characteristics, such as direction and velocity, and give an estimate of intraparenchymal flow resistances. Visualization of the renal vessels can be dramatically improved by using contrast enhancement. In clinical practice, the enhanced US signal provided by contrast agents can reduce the number of technically nondiagnostic cases as well as the number of false-negative results. Furthermore, new technological advances can play an increasing role in renal vascular imaging US. Three-dimensional volume-rendering

Knowledge about the use of different Doppler imaging modalities and typical sonographic findings of the most frequently conditions affecting renal vessels are of great importance. This article reviews the clinical applications of US and Doppler US techniques including basics and technological advances in the field of renal vascular diseases. Keywords Kidney ´ Renal vessels ´ Ultrasound ´ Doppler studies

techniques can be used to recognize shape of vessels with complex course and to analyze relationships among the different structures of the kidney. They can provide panoramic volumetric images of the large abdominal vessels, further outlining the relationships of renal vessels with the abdominal aorta. Blood flow morphology maps can be also displayed within the background of the B-mode data. Combination with contrast agents provides the potential to image the whole vascular renal network and, through special measurement techniques, to estimate vascular volume and transit time, parameters which relate directly to tissue perfusion. The purpose of this paper is to review the clinical applications of US and Doppler US techniques in the field of renal vascular diseases.

Examination technique and normal findings Anatomy The renal arteries (RA) originate from the lateral sides of the aorta, typically at the level of the superior border of the second lumbar vertebra, directed slightly anteri-

1903

orly, approximately 1±1.5 cm below the superior mesenteric artery origin. The right RA branches from the aorta, directs anteriorly for a few centimeters and then descends obliquely, passing posteriorly to the inferior vena cava, in a postero-lateral direction toward the kidney. The left RA has a more horizontal and typically shorter course. It runs from the aorta, posteriorly to the left renal vein, into the renal hilum. Approximately 30 % of patients with normally positioned kidney have multiple renal arteries, with one or more accessory vessels [1, 2, 3]. Most of them arise close to the main RA, although they can originate inferiorly, some distance away, supplying a portion of the lower pole. Because of this variation in anatomy, accessory renal arteries are difficult to detect on Doppler studies, and most are missed. The main RA divides at the hilum, either within or outside of the kidney, into anterior and posterior branches that further divide into segmental and then interlobar arteries. Renal arteries are terminal vessels that do not communicate with each other. They are divided into four vascular regions: apical; anterior; posterior; and inferior [4]. Each segmental branch supplies one of these regions. Approximately 90 % of the normal renal blood flow goes to the cortex; only 10 % supplies the medulla. The interlobar arteries further divide into a network of arcuate arteries that run at the corticomedullary junction and give off the cortical (interlobular) branches, which run radially to the renal periphery, and the medullary branches, which supply the renal pyramids. The renal veins approximately follow the arteries and join at the hilum to form the main renal vein. Intrarenal veins have collaterals, which unite with each other. The right renal vein runs in a postero-anterior direction, with a relatively short course to enter the vena cava. The left vein is more horizontal and crosses anteriorly to the aorta and posteriorly to the superior mesenteric artery to enter the vena cava, at the level of the first lumbar vertebra. It is worth remembering that the left gonadal vein, adrenal vein, and the lumbar veins usually enter the left renal vein.

Transducer position

Fig. 1 Anatomy of the main renal vessels. 1 Transverse mid abdominal section; 2 Oblique longitudinal approach; Ao aorta in a transverse section; IVC inferior vena cava in transverse section; RRV right renal vein; RRA right renal artery; LRV left renal vein; LRA left renal artery

Main renal arteries

Fig. 2 Power Doppler US image of the both normal renal arteries. Abdominal transverse section

The first requirement is to choose a good scan plane. This is a key point. The main renal arteries can be imaged in an abdominal transverse section, by placing the transducer in a midpoint between the xiphoid process and the umbilicus (Fig. 1). By applying compression with the transducer, the bowel loops can be displaced, to see the aorta in a transverse section, together with the origin of both renal arteries. With the axial approach,

the arteries often may be followed to the renal hilum, especially on the right side where the liver can be used as an acoustic window (Fig. 2). In patients with abundant bowel gas obscuring visualization of the aorta, angling the probe cranially or caudally can help to overcome the problem and allow identification of the vessels.

1904

a Fig. 4 Color Doppler imaging of the left renal artery and vein from the left flank using the kidney as an acoustic window

dow to obtain coronal scan planes of the renal hilum and the aorta with the origins of the left renal vessels (Fig. 4).

Normal findings

b Fig. 3 a Color Doppler US image in oblique longitudinal section. Hypoplasia of the right renal artery. b Same patient. Correlative conventional angiography

An alternative method of imaging the origins of RA is the use of an oblique longitudinal approach with the patient in a 45  right anterior oblique position (Fig. 1). The transducer should be placed in the right subcostal position to obtain longitudinal view of the right kidney. Then the probe should be moved in a medial direction following the course of the right renal vein from the hilum to the inferior vena cava. The transducer is angled until the aorta, together with the origins of both RA, appears. The disadvantage of this approach is that only the origins of the vessels are shown. Nevertheless, this projection is best for determining whether accessory arteries, aplasia, or hypoplasia (Fig. 3) of the renal arteries are present. The left renal artery and vein also can be seen from the left flank, using the kidney itself as an acoustic win-

When the origins of renal arteries are imaged with color Doppler in the transverse position, the first segment of the right renal artery has flow directed toward the transducer, then represented in red color. Color change is detected shortly after the origin, where the direction of flow goes posteriorly. Most of the course of the vessels is then displayed in blue. If the origins of renal arteries are imaged in the oblique longitudinal section, right RA passes directly toward the transducer from the aorta and is colored red, whereas left RA courses away from the transducer and is blue (Fig. 5). On color Doppler examination, flow within the renal vein is opposite in color to that within the renal artery. Power Doppler can delineate a better image of the proximal RA without the absence of flow in the arterial segments that run horizontal to the US beam, but with loss of directional and velocity information. Power Doppler provides superb visualization of the entire renal vascular tree from the main RA to the arcuate arteries and beyond. The segmental renal branches lie within the echogenic renal hilum. Interlobar arteries can be visualized lateral to the renal pyramids, and the arcuate vessels run behind the renal pyramids parallel to the renal cortex. With power Doppler several further generations of vessels (the interlobular arteries) are seen radiating to the renal capsule (Fig. 6).

1905

Fig. 5 Color Doppler imaging of the both renal arteries and the aorta. Longitudinal section. Right renal artery shown as red, left renal artery shown as blue

Fig. 6 Power Doppler US image of the right kidney with renal vessels. Good visualization of the entire renal vascular tree

Pulsed Doppler The longitudinal position can be recommended for measuring the Doppler spectrum of both renal arteries due to optimal angle of both RA in this position. The Doppler signal in the RA is a low-resistance signal, similar to that found in all the parenchymal organs of the body. The important features of this signal are the rapid systolic rise and the continuous high-velocity flow throughout diastole (Fig. 7). A small spike occurs at the end of the systolic rise (the so-called early systolic

Fig. 7 Spectral Doppler US image from the right renal artery in normal subjects. Note small spike that occurs at the end of systolic rise. This feature is seen only in the normal main renal artery

peak). It is an important feature to identify when measuring the systolic acceleration. This feature is seen only in the main renal artery and its major branches. The peak systolic velocity (PSV) in the main renal artery and its major branches should be less then 100 cm/s [5]. The velocity slowly decreases in the intrarenal arteries as they branch into the kidney. The diastolic velocity is a little less than half of the systolic velocity. This value normally is expressed as a ratio of end diastole to peak systolic flow, the most commonly used ratio being the resistance index or Pourcelot resistive index (RI), which measures less then 0.7 in the normal kidney [5]. The renal/aortic ratio is used to ªnormalizeº measurements of the velocity within the renal artery. A summary of the most common parameters used to evaluate the flow patterns within the renal arteries is shown in Table 1. The RI values, as measured in healthy subjects, show a significant dependence on age and the area sampled. The values in the main RA are higher in the hilar region (0.65  0.17) than in the more distal small arteries, and they are lowest in the interlobar arteries (0.54  0.20). The RI values are higher in elderly patients. The agedependent RI values are shown in the Table 2. In clinical practice the value of RI 0.7 is used to discriminate between normal and pathologic resistances to flow. Although it is nonspecific, an elevated RI of greater than 0.7 suggests renal parenchymal disease processes or postrenal obstruction [5]. The renal vein Doppler shift signal is continuous in the smaller veins and with slightly phasic changes with respiration in the main veins. The left renal vein shows little or no phasic swing during the cardiac cycle, whereas the right one shows a variable amount of pul-

1906

Table 1 Normal renal Doppler indices. S peak systolic velocity; D end diastolic velocity. (From [6]) Index

Formula

Normal value

S±D/S

60±100 sm/s (< 180 sm/s) 0.56±0.7 (< 0.7)

S: main renal artery Resistance index Pulsatility index

S±D/mean

0.7±0.14

Renal/aortic ratio

Renal systolic velocity/ aortic systolic velocity

< 3.0

Systolic rise time

Time to early systolic peak

0.11  0.06 11  8 m/s2

Systolic acceleration

Table 2 Normal resistive index values in the interlobar arteries according to patient age. (From [7]) Age (years)

Mean

Mean  2 SD

< 20 21±30 31±40 41±50 51±60 61±70 71±80 > 80

0.567 0.573 0.588 0.618 0.688 0.732 0.781 0.832

0.523±0.611 0.528±0.618 0.546±0.630 0.561±0.675 0.603±0.733 0.649±0.815 0.707±0.855 ±

satility, reflecting changes in right atrial pressure. This can be related to its shorter course, and to the common absence of valvular structures within it (Fig. 8). The rate of technically inadequate color Doppler US examinations of the main RA varies between 9 % [8] and 23.5 % [9]. Some optimistic reports, such as those from Robertson et al. [10], who carefully controlled the duration of the examination limiting attempts to 20 min (allowing a longer time was impractical in that clinical setting) found an identification rate of 95.5 % of right and 82 % of left renal arteries; however, others, such as Berland et al. [11], are less optimistic. From their study with clinical comparisons with angiography, duplex scanning showed unsatisfactory results: Only 58 % of the main RA and no accessory RA were found [11]. In this way, contrast agents should have application in cases where the Doppler signal is difficult to obtain, either because of a weak signal or because of signal attenuation by overlying tissue [12]. The enhanced signal provided by the contrast agent should reduce the number of technically nondiagnostic cases and the number of false-negative results.

Renal artery stenosis Renal artery stenosis (RAS) is a relatively rare but important cause of renovascular hypertension. Both arte-

Fig. 8 Spectral Doppler US image of the left renal vein in normal subjects. Spectral trace shows little swing within the cardiac cycle

riosclerosis, and less commonly fibromuscular dysplasia, can lead to renovascular hypertension. Atheromatous lesions involve the proximal renal artery, whereas fibromuscular dysplasia involves the distal main renal artery and segmental renal arteries. Detection of RAS is important since it is a potentially curable cause of hypertension, by the use of radiologic techniques or surgery, and is a direct contraindication to the use of angiotensin converting enzyme inhibitors during medical therapy of hypertension. Conventional angiography is currently the gold standard in the detection of renal artery stenosis. Unfortunately, it is an invasive test and is not suitable as a screening procedure. Magnetic resonance angiography is a very promising modality in this field and, albeit expensive, is non-invasive but cannot be used as a screening, widely available test [13]. Contrastenhanced helical CT is a promising test also, especially when multidetector equipment is used. The need of iodinated contrast medium, however, does not allow its widespread use as a screening test, given possible nephrotoxicity. Doppler US is a non-invasive, widespread and relatively inexpensive diagnostic modality. It is not surprising, therefore, that it has been extensively investigated as a screening test for renal artery stenosis (Fig. 9) [14, 15]. Duplex US criteria of RAS can be divided into two groups based on direct findings obtained at the level of the stenosis (proximal criteria), or on flow changes observed in the renal vasculature, distal to the site of stenosis (distal criteria). The proximal criteria are direct signs obtained at the site of the stenosis. The first, most important sign is the increase in peak systolic velocities (PSV). Velocities higher than 1.5 m/s are significant and should be used as

1907

a Fig. 10 Spectral Doppler waveform from the stenotic area in the right RA. Increased peak systolic velocities are seen

b Fig. 9 a Power Doppler US image of the right renal artery with echogenic plaque inside the vessel close to the origin. b Same patient. Correlative conventional angiography with the plaque in the origin of the right RA Fig. 11 Color Doppler imaging of the right RAS. Mosaic flow is seen within the stenotic area

a criterion for RAS 50 %, and > 1.8 m/s for stenosis 60 % (Fig. 10). Comparison of the velocities in the aorta with those in the renal arteries, the so-called renal/aortic ratio (RAR), is also helpful. A PSV in the renal artery three times higher than the aortic PSV indicates the presence of RAS [16]. The use of the ratio (RAR) instead of the absolute peak systolic value is preferable since hypertension itself can cause an increase of peak systolic flow velocities within all vessels of the hypertensive patient. The second criterion is the presence of poststenotic turbulences; these are seen as a widening of the Doppler trace at spectral analysis of signals at the stenosis and as a mosaic color pattern on the color Doppler image (Fig. 11).

Distal criteria analyze the flow changes induced by the stenosis at the level of intrarenal vessels. Patients with severe renal artery stenosis commonly present with pathologic intrarenal signals, the so-called tardus-parvus waveform, first described by Handa et al. [17]. Tardus means slow and late and parvus means small and little. Tardus refers to the fact that systolic acceleration of the waveform is slowed, with consequent increase in time to reach the systolic peak. Parvus refers to the fact that the systolic peak is of low height, indicating slowed velocity (Fig. 12). Poststenotic systolic peak are rounded with lengthened systolic rise time (or slow systolic

1908

a

b Fig. 12 a Normal renal artery waveform. The acceleration index (AI) is the slope of the line (m/s2) connecting the two points representing onset of systole and the early systolic peak complex. The acceleration time (AT) is the time in seconds between those two points. b A tardus-parvus renal artery waveform. Rounding and flattening of the systolic peak and prolongation of the AT. The AI is more horizontal

acceleration time ± the time in seconds from the onset of the systole to peak systole), slower than 0.07 s. The acceleration index (the slope of systolic upstroke) is decreased < 3m/s2 [18]. Although this phenomenon has been postulated to be secondary to decreased perfusion pressure [19], the explanation actually is more complicated, as systolic acceleration also relates to peripheral resistance, vessel compliance, and other variables, as well as to upstream stenosis. For example, it has been shown that increasing vessel compliance accentuates the decrease in systolic acceleration independent of the transstenotic pressure drop [20, 21]. Although generally confirming the efficiency of tardus parvus, Stavros et al. pointed out that simple pattern recognition of Doppler tracings from segmental renal

arteries was more valuable than calculating the acceleration index and acceleration time, with loss of the normally seen early systolic peak indicating RAS [22]. Persistence of the early systolic peak could be observed only in patients with mild stenosis. In their study an acceleration index less than 3.0 m/s2 had an accuracy of 85 %, a sensitivity of 89 %, and specificity of 83 % for detecting RAS. An acceleration time of ³0.07 s was 89 % accurate, 78 % sensitive, and 94 % specific. The tardus-parvus waveform was 96 % accurate, 95 % sensitive, and 97 % specific for RAS [22]. Sensitivity of the technique may be improved by the administration of Captopril. Rene et al. [23] studied 62 renal arteries in 31 hypertensive patients who underwent Doppler scanning before and 1 h after administration of Captopril prior to angiography. They concluded that abnormalities of the tardus-parvus phenomenon of RAS in distal vessels can be made more apparent after administration of Captopril with an improvement of sensitivity. Many studies have shown that analysis of distal signs observed at Doppler US can be a useful fist approach to patients with suspected RAS [22, 24, 25]. Other reports found recognition of tardus-parvus effect unsatisfactory [26]. Many factors influence systolic acceleration and may make the test non-specific. Extrarenal factors, such as aortic/mitral valvular disease, left ventricular dysfunction, or even cardiovascular medications, might affect systolic acceleration as well. Numerous factors, such as age, hypertension, and diabetes, affect vessel compliance. Such variables may explain why some authors have not been able to reproduce these results [26]. Additional distal criteria have been developed: A study by Schwerk et al. demonstrated that differences in resistive index (RI) could help in diagnosing RAS [27]. Decreasing PSV in RAS results in lowering of resistive index (RI) values. Right-to-left difference DRI between kidneys greater than 0.05 had a sensitivity of 100 % for RAS greater than 60 %. Halpern et al. suggested that patients should be screened by distal measurements of early systolic acceleration, because determining RAR when segmental samplings are normal would be superfluous [28]. Using combined criteria (Table 3) pertaining to the stenosis site and downstream patterns the sensitivity of Doppler US varies from 64 to 89 %, and specificity from 82 to 99 % [9, 15, 25]. In renal transplants, which are easier to explore because of superficial location, as well as of knowledge of the course of the renal artery from the surgical report, the results were more convincing. Although Doppler techniques are well suited as a screening test in transplanted kidneys, their use for suspected renal artery stenosis in native kidneys has some problems. The main problem is that it is difficult to visualize the whole of both renal arteries in all patients [11]. Moreover, acces-

1909

Table 3 Criteria for RAS. (From [29]). PSV peak systolic velocity; RAR renal/aortic ratio; AT acceleration time; AI acceleration; RI resistive index Main RA PSV in stenotic area RAR Intrarenal arteries ESP AT AI RI DRI (left±right)

> 1.8 m/s > 3.0 Absent > 0.07 s < 3 m/s > 0.8 > 5%

Doppler signals over time after intravenous injection of a bolus of contrast medium. This technique produces time±intensity curves, which, in renal artery stenosis, have area under the curve larger than in normal kidneys [34]. In severe stenosis, furthermore, there is also a delay in the wash-in phase of the curve. At present, however, only preliminary results have been presented in the literature, and further studies are needed before the introduction of this technique in clinical practice.

Renal vein thrombosis sory renal arteries are usually missed on Doppler studies, and stenosis of these may also cause hypertension. Also, Doppler studies may be insensitive in patients with mild stenosis; thus, the role of Doppler sonography as a screening test in hypertensive patients remains controversial. At present, its use cannot be separated from a careful clinical evaluation of the patient population. Given its technical difficulties, it has to be employed in patients well selected based on clinical criteria of high probability of having RAS. Ultrasound contrast agents have recently added new possibilities to color and duplex Doppler in the detection of RAS [30, 31]. They have been shown to increase the percentage of diagnostic examinations in analyses of main renal arteries [30]. With echo enhancers a renewal of interest in Doppler studies of the main renal arteries is occurring. Ultrasound contrast agents increase the intensity of the Doppler signals, thus producing more rapid and complete visualization of the intrarenal and extrarenal arteries [29]. Contrast agents have application in cases where the Doppler signal is difficult to obtain, either because of signal attenuation by overlying tissue or because of a weak signal [12]. Missouris et al. [32] suggest that renal duplex scanning using contrast enhancement is a promising new non-invasive technique in screening patients with suspected RAS. Contrast enhancement produces more reproducible spectral waveforms, improves accuracy, and halves the examination time [32]. A recent study by Claudon et al. [33] showed that the number of examinations with successful results increased following enhanced Doppler US examination compared with nonenhanced Doppler US, including patients with obesity or renal dysfunction. Moreover, the agreement between US data and angiography in RAS was higher with enhanced Doppler US. They have shown that contrast media decrease examination time but do not cause an increase in sensitivity [33]. Contrast media for US do not undergo renal filtration or tubular excretion and can be, on the whole, considered as purely vascular tracers. A new interesting application of these agents in suspected RAS is quantification of the renal enhancement of color or power

Renal vein thrombosis may occur in up to 40 % in dehydrated or septic infants [35]. In native kidneys, renal vein thrombosis starts in small intrarenal veins in situations of faulty coagulation mechanism and slowed flow [36]. The post-glomerular circulation, because of slow flow, is particularly prone to thrombosis. In adults it most commonly appears in association with renal disease including glomerulonephritis, systemic lupus erythematosus, diabetus mellitus, nephrotic syndrome, in severe hypovolemic shock, and following kidney transplantation. The US features are non-specific, and only renal enlargement, with decreased echogenicity in the early stages followed by an increase in cortical reflectivity can be detected [37]. Doppler US cannot accurately diagnose thromboses of intraparenchymal veins. The presence of venous Doppler signals within the kidney or renal vein, in fact, does not exclude the diagnosis of a thrombosis involving only one of the many intraparenchymal veins. In fact, although in acute renal vein thrombosis, the whole kidney is underperfused, and although only arterial signals can be seen at the renal hilum, it must be remembered that venous collaterals develop rapidly, and when this happens venous signals are re-established. Then, the presence of parenchymal venous flow does not exclude RVT, as collateral flow develops very quickly, particularly in children (Fig. 13) [38]. Color Doppler can be accurate for diagnosing chronic renal vein thrombosis when the main renal vein can be directly visualized, and flow signals cannot be detected in it [38]. Renal vein thrombosis can be caused also from tumor involvement in patients with renal cell carcinoma (RCC). Color Doppler sonography is accurate in demonstrating tumor thrombus in the renal veins. The US vascular features include distension of the renal vein, full of echogenic material (Fig. 14). The presence of arterial Doppler signals within the thrombus allows unequivocal demonstration of tumor involvement of the vessel. In patients with renal transplant, with complete thrombosis of the veins of the allograft, reduction of diastolic flow in RA and reversal of flow in diastole with distended renal vein and absence of flow signals from the renal vein have been reported as pathognomonic signs of renal vein thrombosis [39].

1910

Fig. 13 Chronic thrombosis of the right renal vein. Absence of flow in the main right renal vein. Collateral flow is clearly seen

Fig. 15 Color Doppler imaging of the abdominal infrarenal aneurysm. Aneurysm arises below the origins of both renal arteries

Aneurysms

Fig. 14 Power Doppler US image of the thrombosed right renal vein

In very lean, but otherwise healthy subjects, the left renal vein can become compressed between the aorta and the superior mesenteric artery, resulting in the socalled nutcracker syndrome (renal vein entrapment syndrome) [40]. Kim et al. proposed that a cut-off value of greater than 5.0 for the ratio of antero-posterior diameter and the ratio of peak velocity (both the AP diameter and PV being measured at hilar and aorto-mesenteric sites of the left renal vein) be used as a criterion in diagnosing nutcracker syndrome [41].

Aneurysms due to atherosclerosis usually occur in the infrarenal aorta and common iliac arteries; however, they are also found in the renal arteries. Most renal artery aneurysms have been found in persons 50±70 years of age. Renal artery aneurysms can cause rupture, thrombosis, embolization, and dissection [42]. Color Doppler US provides an effective, non-invasive means of diagnosing renal artery aneurysm. Aneurysms may be identified along the course of the main renal artery as an outpouching containing color flow. Slow velocities and a whirling pattern of flow can usually be observed on Doppler studies. Color Doppler US can provide a quick, easy way in demonstration of aneurysmal dilatations of the vascular wall in the abdominal aorta. Most fusiform and saccular aneurysms of the aorta arise below the level of renal arteries, but it is nevertheless important to determine whether the renal vessels are involved, since this alters patient management. Aortic dissection may extend into a renal artery, thus interrupting renal blood flow. Scanning along the longitudinal approach is the best way to demonstrate the relationships between the aneurysm and RA (Fig. 15).

Arteriovenous fistulas Both congenital and postbiopsy renal AV fistulae can be diagnosed on color Doppler examination. The most frequent cause of AV fistula in both the native and transplanted kidney is complication of percutaneous biopsy [43]. Small fistulas are not visible by conventional US,

1911

a Fig. 16 Arteriovenous postbiopsy fistula of the kidney. Spectral Doppler waveform shows low-resistance afferent artery spectrum. (Courtesy of L. E. Derchi, Genova)

and can be detected by color Doppler only. The AV fistula often appears on color Doppler US as a non-specific mosaic pattern with color aliasing, reflecting rapid flow rate and fine movements of tissues surrounding it [44]. Two adjacent vessels can be usually recognized in which waveform analysis shows decreased RI and increased peak flow velocities in the afferent artery, and pulsatile flow pattern in the efferent vein (Fig. 16). Steal phenomena can case hypoperfusion of renal tissues surrounding the fistula [43]. Large AV fistulas are seen as cystic or complex structures in which color Doppler demonstrates vortices of high-velocity, low-impedance flow. Although rare, AV fistulas are potentially lethal pathologic conditions. Failure to recognize properly an AV fistula, in fact, puts the patient at high risk of hemorrhagic complications during a renal biopsy.

b Fig. 17 a, b Patient 12 years after transplantation of the kidney. Renal allograft dysfunction. Chronic rejection. a Power Doppler of the allograft. Marked decreasing of the cortical vascularity. b A 3D Power Doppler angiography shows reduced density and tortuosity of the interlobular vessels

Renal allografts The examination technique for renal allografts is much easier than that for native kidneys, due to the superficial location of the transplanted organ. The examination of the transplanted kidney should be performed using superficial 7.5-MHz probes to study cortical perfusion and abdominal convex 3.5-MHz probes for direct visualization of the deeper structures such as transplanted artery and vein. The allograft vascular imaging protocol should include visualization of the anastomotic region to exclude vascular stenosis, anastomotic aneurysms, or false aneurysms. Also accurate detection of the course and flow in the transplanted artery and vein should be performed

due to possible stenosis and occlusions. Using superficial probes and Power Doppler mode, flow within the cortical region of the whole kidney should be studied. Intrarenal resistance indices (RI, PI) should be sampled. Arteriovenous fistula and false aneurysms could be found after biopsy in transplanted kidney In normal transplants the values of RI is less than 0.71, and shows a slight decrease toward the periphery. A reduced diastolic flow velocity associated with an increase in intrarenal resistance index can be detected in acute and chronic rejection reactions, urinary obstruction, arteriosclerosis of the vasculature, and acute tubular necrosis [45, 46, 47]. The significance of the changes in RI

1912

and the diagnostic predictive value of these indices in assessing the etiology of the transplant-related complications remain controversial [48, 49]. Using superficial probes a tiny blush through the whole cortical region up to the capsule should be seen in a normally functioning renal allograft [50]. According to Martinoli et al. [60], loss of visualization of interlobular vessels at Power Doppler US could be regarded as an additional hallmark of renal transplant dysfunction. But abnormalities of the interlobular vascular pattern in some patients with chronic rejection are probably nonspecific in identifying the nature of renal transplant dysfunction (Fig. 17). An anastomotic stenosis could be assumed when the registered maximum systolic velocities exceed 150±200 sm/s, shows a local increase of more than 50 %, or if the velocity related to the iliac artery increases beyond a factor of 2.6±3.0 [51, 52, 53]. In a study of 109 transplanted kidneys, a peak systolic velocity ³2.5sm/s in the transplanted renal artery had a sensitivity of 100 % and specificity 95 % for the detection of RAS, although the use of measurements of RI, acceleration index, and time in intrarenal vessels were less useful as discriminating diagnostic tests between normal and stenosed group of patients with transplants [54]. Occlusions of the transplanted or segmental arteries show complete or regionally limited flow in the intrarenal vasculature [45, 55]. Absence of flow with pulsed and color Doppler is a valuable additional sign confirming the diagnosis [56]. Arteriovenous fistulas can develop after percutaneous biopsies of kidney transplants. They are recognized by focal massively disturbed flow, with high intrarenal flow velocities [43]. Renal venous thrombosis can be confirmed by absence of venous flow intrarenally and in the renal vein, as well as on the basis of high-impedance arterial Doppler signals, which present reversed flow during the diastole [57].

Fig. 18 Ultrasound contrast study of renal perfusion. Flash Echo imaging. (Provided by Toshiba Medical Systems Europe)

Technical advances

Fig. 19 Three-dimensional US angiography of the both renal arteries and the entire aorta

Modern commercially available machines are now able to outline with great detail the intraparenchymal vasculature of the kidney. Studies of cortical perfusion are possible using relatively high-frequency (at least 5 MHz) high-resolution transducers and preferably using Power Doppler. The use of Power Doppler with contrast agents not only facilitates this study but enables perfusion studies ± similar to those created by isotope studies ± to be performed [58]. Signals from interlobular vessels are visible in the whole cortex, including the most peripheral region close to the capsule. Harmonic imaging with contrast agent injections can be used to map regional differences in flow as well as quantitative measurements of a contrast agent's transit time and has the potential to as-

sess kidney abnormalities associated with renal blood flow [59]. Studies of cortical perfusion are new, and therefore only preliminary experiences have been reported in the literature [56, 60, 61]; however, the first results are very promising in establishing a pathologic±sonographic correlation in acute renal parenchymal inflammation [62] and in renal allograft evaluation [56, 60, 61]. It must be remembered that care should be taken in diagnoses of perfusion defects, since absence of detectable flow at the interlobular level does not always correspond to cortical areas that lack perfusion on other, more reliable techniques, such as MR [60].

1913

a

a

b Fig. 21 a A 3D US angiography of the right renal arteries and aorta. Accessory renal artery is clearly seen. b Same patient: correlative conventional angiography

b Fig. 20 a A 3D US angiography of the right kidney. Maximum intensity projection image demonstrates accessory renal artery. b A 3D volume-rendering image of the left kidney. Accessory left renal artery

In clinical practice, the enhanced US signal provided by contrast agents can reduce the number of technically nondiagnostic cases, as well as the number of false-negative results. With additional use of contrast agents and phase inversion harmonic imaging the whole course of the renal arteries could be imaged without motion or aliasing artifacts. Harmonic ultrasound with contrast agents added blood flow morphology maps within the background of

the B-mode volumetric data. Combination with contrast agents provides the potential to image the whole vascular renal network and, through special measurement technique (Flash Echo), to estimate vascular volume and transit time, parameters which relate directly to tissue perfusion (Fig. 18). One of the promising and rapidly developing techniques is three-dimensional US angiography [63, 64, 65, 66, 67]. Three-dimensional US can overcome some drawbacks of 2D US and can provide the angiogram-like images of both, renal arteries and the entire aorta (Fig. 19). Careful freehand scanning with a smooth, linear translation, or sector sweep, can acquire 3D data sets in a single breath-hold. Different approaches can be used to acquire 3D data sets of renal vessel: anterior or ante-

1914

rolateral, for evaluation of both RA and entire aorta; and coronal for visualization of renal vasculature, main RA, and entire aorta. After acquisition of 3D data, postprocessing is performed using maximum and minimum intensity projections (MIP or MinIP) to obtain angiogramlike 3D images for further analysis (Fig. 20). We believe that 3D US angiography has the potential to become excellent for screening in evaluation of accessory renal arteries especially in children, young adults, patients with renal failure, allergy to iodinated contrast agents, fear of ionizing radiation, or arterial catheterization. Magnetic resonance angiography, with its high cost and less availability, should be reserved for problem cases. It is known that accessory renal arteries are found frequently. Previous researchers have reported poor ability of color Doppler US to depict accessory renal arteries [15]. Contrast agents and harmonic imaging may have a role to play in future and increase the sensitivity and specificity of color Doppler US in detection

of accessory renal arteries [68]. Three-dimensional US angiography can enhance the possibility of 2D US in evaluation of accessory renal arteries. An angiogramlike image of 3D US angiography is becoming an excellent alternative to conventional angiography (Fig. 21). This method is promising as the primary study for accessory RA and for determining the relationships of the origins of the RA to abdominal aneurysms, and can be the preferred technique for patients with a contraindication to conventional angiography. Further developments of computing power, widespread diffusion of state-of-the-art US machines, matrix-array transducers, harmonic imaging techniques and reconstruction algorithms for surface and volume rendering will lead real-time 3D US scans to become routine in clinical practice. Acknowledgements I am grateful for all the help provided by L. Derchi during preparation of this article.

References 1. Bakker J, Beek FJ, Beutler JJ et al. (1998) Renal artery stenosis and accessory renal arteries: accuracy of detection and visualization with gadoliniumenhanced breath-hold MR angiography. Radiology 207: 497±504 2. Korst MB, Joosten FB, Postma CT et al. (2000) Accuracy of normal-dose Contrast-enhanced MR angiography in assessing renal artery stenosis and accessory renal artery stenosis and accessory renal arteries. Am J Roentgenol 174: 629±634 3. Neri E, Caramella D, Bisogni C et al. (1999) Detection of accessory renal arteries with virtual vascular endoscopy of the aorta. Cardiovasc Intervent Radiol 22: 1±6 4. Beregi J-P, Elkohen M, Deklunder G et al. (1996) Helical CT angiography compared with arteriography in the detection of renal artery stenosis. Am J Roentgenol 167: 495±501 5. Platt JF, Rubin JM, Ellis JH (1989) Distinction between obstructive and nonobstructive pyelocaliectasis with duplex Doppler sonography. Am J Roentgenol 153: 997±1000 6. McGahan JP, Goldberg BB (1998) Diagnostic ultrasound. A logical approach. Lippincott Raven, Philadelphia, pp 1288: 793 7. Krumme B, Kirschner T, Gondolf D et al. (1994) Altersabhanigkeit des intrarenalen Resistance Index (RI) bei essentiellen Hypertonikern. Bildgebung Imaging (Suppl) 2: 55

8. Zoller WG, Hermans H, Bogner JR et al. (1990) Duplex sonography in the diagnosis of renovascular hypertension. Klin Wochenschr 68: 830±834 9. Postma CT, van Aalen J, de Boo T et al. (1992) Doppler US scanning in the detection of renal artery stenosis in hypertensive patients. Br J Radiol 65: 857±860 10. Robertson R, Murphy A, Dubbins PA (1988) Renal artery stenosis: the use of duplex ultrasound as a screening technique. Br J Radiol 61: 196±201 11. Berland LL, Koslin DB, Routh WD et al. (1990) Renal artery stenosis: prospective evaluation of diagnosis with color duplex US compared with angiography. Work in progress. Radiology 174: 421±424 12. Cosgrove D (1997) Why do we need contrast agents for ultrasound? Clin Radiol 51 (Suppl):1±4 13. Debatin JF, Spritzer CE, Grist TM et al (1991) Imaging of the renal arteries: value of MR angiography. Am J Roentgenol 157: 981±990 14. Soulez G, Oliva V, Turpin S et al. (2000) Imaging of renovascular hypertension: respective values of renal scintigraphy, renal Doppler US, and MR angiography. Radiographics 20: 1355±1368 15. Cobelli F de, Venturini M, Vanzulli A et al. (2000) Renal arterial stenosis: prospective comparison of color Doppler ultrasound and breath-hold, threedimensional, dynamic, gadolinium-enhanced MR angiography. Radiology 214: 373±380

16. House MK, Dowling RJ, King P et al. (1999) Using Doppler sonography to reveal renal artery stenosis: an evaluation of the optimal imaging parameters. Am J Roentgenol 173: 761±765 17. Handa N, Fukunaga R, Uehara A et al. (1986) Echo-Doppler velocimeter in the diagnosis of hypertensive patients: the renal artery Doppler technique. Ultrasound Med Biol 12: 945±952 18. Patriquin HB, Lafortune M, Jequeier J et al. (1992) Stenosis of the renal artery: assessment with Doppler sonography. Radiology 184: 479±485 19. Lafortune M, Patriquin H, Demeule E et al. (1992) Renal arterial stenosis: slowed systole in the downstream circulation: experimental study in dogs. Radiology 184: 475±478 20. Bude RO, Rubin JM, Platt JF et al. (1990) Pulsus tardus: its cause and potential limitations in detection of arterial stenosis. Radiology 134: 779±784 21. Bude RO, Rubin JM (1995) Detection of renal artery stenosis with Doppler sonography: it is more complicated than originally thought. Radiology 196: 612±613 22. Stavros AT, Parker SH, Yakes WF, et al. (1992) Segmental stenosis of the renal artery: pattern recognition of tardus and parvus abnormalities with duplex sonography. Radiology 184: 487±492 23. Rene PC, Oliva VL, Bui BT et al. (1995) Renal artery stenosis: evaluation of Doppler US after inhibition of angiotensin-converting enzyme with Captopril. Radiology 196: 675±679

1915

24. Kaplan-Pavlovic S, Nadja C (1998) Captopril renography and duplex Doppler sonography in the diagnosis of renovascular hypertension. Nephrol Dial Transplant 13: 313±317 25. Krumme B, Blum U, Schwertferger E et al. (1996) Diagnosis of renovascular disease by intra- and extrarenal Doppler scanning. Kidney Int 50: 1288±1296 26. Kliewer MA, Tupler RH, Carroll BA et al. (1993) Renal artery stenosis: analysis of Doppler waveform parameters and tardus-parvus pattern. Radiology 189: 779±787 27. Schwerk WB, Restrepo IK, Stellwaag M et al. (1994) Renal artery stenosis grading with image-directed Doppler US evaluation of renal resistive index. Radiology 190: 785±790 28. Halpern EJ, Deane CR, Needleman L et al. (1995) Normal renal artery spectral Doppler waveform: a closer look. Radiology 196: 667±673 29. Lavopierre AM, Dowling RJ, Little AF et al. (2000) Ultrasound of the renal vasculature. Ultrasound Quarterly 16: 123±132 30. Melany ML, Grant EG (1997) Clinical experience with sonographic contrast agents. Semin Ultrasound CT MR 18: 3±12 31. Dowling RJ, House MK, King PM et al. (1999) Contrast-enhanced Doppler ultrasound for renal artery stenosis. Australas Radiol 43: 206±209 32. Missouris CG, Allen MC, Balen FG et al. (1996) Non-invasive screening for renal artery stenosis with ultrasound contrast enhancement. J Hypertens 14: 519±524 33. Claudon M, Plouin PF, Baxter GM et al. (2000) Renal arteries in patients at risk of renal arterial stenosis: multicenter evaluation of the echo-enhancer SH U 508A at Color an spectral Doppler US. Radiology 214: 739±746 34. Lencioni RA, Pinto S, Napoli V et al. (1999) Detection of renal artery stenosis by time±intensity analysis of renal enhancement curve at harmonic power Doppler imaging: a pilot clinical study. Radiology 213: 363±364 35. Clark RA, Colley DP (1980) Radiological evaluation of renal vein thrombosis. CRC Crit Rev Diagn Imaging 13: 337±344 36. Gonzales R, Schwarts S, Sheldon C et al. (1982) Bilateral renal vein thrombosis in infancy and childhood. Urol Clin Am 9: 279±283 37. Scoutt LM, Brown JM, Hammers LW (1997) Color Doppler evaluation of the native kidney. Appl Radiol:9±23 38. Rosenfeld AT, Zeeman RK, Kronen JJ et al. (1980) Ultrasound in experimental and clinical renal vein thrombosis. Radiology 137: 735

39. Baxter GM, Morley P, Dall B (1991) Acute renal vein thrombosis in renal allografts: new Doppler ultrasonic findings. Clin Radiol 43: 125 40. Wendel RG, Crawford ED, Hehman KN et al. (1980) The nutcracker phenomenon: an unusual cause for renal bleeding of unknown origin. J Urol 123: 761±763 41. Kim SH, Cho SW, Kim HD et al. (1996) Nutcracker syndrome: diagnosis with Doppler US. Radiology 198: 93±97 42. Dong Q, Schoenberg SO, Carlos RC et al. (1999) Diagnosis of renal vascular disease with MR Angiography. Radiographics 129: 1535±1554 43. Middleton WD, Kellman GM, Melson GL et al. (1989) Postbiopsy renal transplant arteriovenous fistulas: color Doppler US characteristics. Radiology 171: 253±257 44. Takebayashi S, Aida N, Matsui K (1991) Arteriovenous malformations of the kidneys: diagnosis and follow-up with color Doppler sonography in 6 patients. Am J Roentgenol 157: 991±995 45. Taylor KJW, Morse SS, Rigsby CM (1987) Vascular complications in renal allografts. Detection with Dupplex Doppler US. Radiology 162: 31±38 46. Harris DC, Antico V, Allen S et al. (1989) Doppler assessment in renal transplantation. Transplant Proc 21: 1895±1896 47. Fluckiger F, Steiner S, Horn M et al. (1990) Farbkodierte Dupplexsonographie und Widerstandsindex bei Nierentransplantation mit Dysfunktion. Fortschr Rontgenstr 153: 692±697 48. Mallek R, Mostbeck G, Kain R et al. (1990) Vaskulare Nierentransplantatabstossung ± Ist eine duplexsonographische Diagnose moglich? Fortschr Rontgenstr 152: 283±286 49. Meyer M, Paushter D, Steinmuller D (1990) The use of duplex Doppler ultrasonography to evaluate renal allograft dysfunction. Transplantation 50: 974±978 50. Claudon M, Blum AG, Martin Bertaux A et al. (1995) Kidney transplantation follow-up: value of power Doppler sonography. Radiology 197: 496 51. Leichtman A, Sorrell K, Wombolt D et al. (1989) Duplex imaging of the renal transplant. Transplant Proc 21: 3607±3610 52. Deane C, Cairns H, Walters H et al. (1990) Diagnosis of renal transplant artery stenosis by color Doppler ultrasonography. Transplant Proc 22: 1395 53. Alvarez G, Gonzalez-Molina M, Cabello M et al. (1991) Pulsed and continuous Doppler evaluation of renal dysfunction after kidney transplantation. Eur J Radiol 12: 108±112

54. Baxter GM, Ireland H, Moss J et al. (1995) Color Doppler US in renal transplant artery stenosis: which Doppler index? Clin Radiol 50: 618±622 55. Grenier N, Douws C, Morel D et al. (1991) Detection of vascular complications in renal allografts with color Doppler flow imaging. Radiology 178: 217±223 56. Trillaud H, Merville P, Linh PTL et al. (1998) Color Doppler sonography in early renal transplantation follow-up: resistive index measurements versus power Doppler sonography. Am J Roentgenol 171: 1611±1615 57. Krumme B (1994) Farbkodierte Duplexsonographie in der Diagnostic von Nierenarterienstenosen nach allogener Nierentransplantation. In: Keller E, Krumme B (eds) Farbkodierte Duplexsonographie in der Nephrologie. Springer, Berlin Heidelberg New York 58. Huber S, Steinbach R, Sommer O et al. (2000) Contrast-enhanced power Doppler harmonic imaging: influence on visualization of renal vasculature. Ultrasound Med Biol 26: 1109±1115 59. Sehgal CM, Arger PH, Pugh CR et al. (1998) Comparison of power Doppler and B-scan sonography for renal imaging using a sonographic contrast agent. J Ultrasound Med 17: 751±756 60. Martinoli C, Crespi G, Bertolotto M et al. (1996) Interlobular vasculature in renal transplants: a Power Doppler US study with MR correlation. Radiology 200: 111±117 61. Claudon M, Grenier N (1997) Technology advances in renal sonography. Diagn Imaging Europe:31±37 62. Clautice-Engle T, Jeffrey R (1997) Renal hypoperfusion: value of power Doppler imaging. Am J Roentgenol 168: 1227±1231 63. Keberle M, Jenett M, Beissert M et al. (2000) Three-dimensional power Doppler sonography in screening for carotid artery disease. J Clin Ultrasound 28: 441±451 64. Lees W (1999) Three- and 4-dimensional ultrasound imaging. Med Mundi l43: 23±30 65. Merz E (1997) Current technical possibilities of 3D ultrasound in gynaecology and obstetrics. Ultraschall Med 18: 190±195 66. Shields LE, Lowery C, Deforge C et al. (1998) 3-Scape real time 3D imaging for ultrasound. Electromedica 66: 84±88 Ä (2000) 3D expands hori67. Weismann N zons in daily clinical practice. Diagn Imaging (Suppl):12±15 68. Sandrick K (2000) 3D ultrasound: more than just a pretty picture. Diagn Imaging (Suppl):2±8