Renal Vascular Doppler Ultrasound

[sonoedu]

The Kidneys

Therese M. Weber, Michelle L. Robbin and Mark E. Lockhart

Renal Vascular Doppler Ultrasound

Ultrasound is the imaging modality of choice for eval‎uation of the kidneys, especially in patients with borderline renal function, the incidence of which is increasing. In comparison with other modalities, ultrasound has the distinct advantage of providing clinically diagnostic information without the need for ionising radiation or contrast agents. The combination of spectral, colour, and/or power Doppler is extremely helpful in renal vasculature eval‎uation.

CLINICAL CONSIDERATIONS

Common clinical indications for renal ultrasound include renal insufficiency and renal failure. Specifically, the request to exclude renal obstruction as the aetiology of acute renal failure leads the list. Doppler ultrasound is not routinely performed to eval‎uate acute renal failure but may be prompted by certain clinical indicators (Box 9-1) or greyscale findings. Colour and spectral Doppler is more commonly used in the native kidneys for eval‎uation of unexplained or uncontrolled hypertension caused by renal artery stenosis (RAS) or for determination of vessel patency. In hypertensive patients, some authors suggest Doppler should be reserved for those patients with a strong clinical suspicion for RAS who are likely to benefit from intervention.¹ As will be shown, there are many other vascular abnormalities than can be demonstrated by Doppler, and these can present with a wide variety of symptoms or signs.

BOX 9-1   CLINICAL CRITERIA USED TO SELECT WHO SHOULD BE EVAL‎UATED FOR RENAL ARTERY STENOSIS (RAS)

• Hypertension with clinical concern for renal artery stenosis

• Hypertension uncontrolled by medical therapy

• Hypertension with abrupt onset

• Hypertension after ACE inhibitor

• Hypertension with discrepant renal size on greyscale US

• Audible abdominal bruit

• Patient with aortic aneurysm

• Patient with aortic dissection

• Patient with renal insufficiency at clinical risk for RAS

• Follow-up of known RAS after vascular therapy

TECHNICAL CONSIDERATIONS

Renal Doppler ultrasound can be one of the most challenging vascular ultrasound examinations. Extensive training and experience in performance and interpretation of the study will reduce examination time, improve study quality, and optimise diagnostic accuracy. A dedicated quality control programme is an important mechanism to assess accuracy and to review cases in which an incorrect diagnosis was made. The goal of this review process should be to improve the quality of future studies.

One common challenge in renal vascular ultrasound is direct visualisation of the proximal renal arteries. Overlying bowel gas may completely obscure the renal vascular origins and result in a non-diagnostic study. To improve the likelihood of a diagnostic study, we request that patients be fasting for at least 6 to 8 hours, when possible, to decrease bowel gas. Another limitation is the deep location of the native renal vessels, especially in obese patients, and utilisation of appropriate sonographic windows may be helpful. Graded transducer pressure can bring the transducer closer to the arteries and simultaneously displace overlying gas. However, the renal arteries occasionally may not be directly visualised despite optimal technique; in some of these technically difficult cases, the identification of segmental arterial waveform abnormalities may still allow successful diagnosis of RAS.

Main Renal Artery Eval‎uation

Doppler eval‎uation of the renal arteries should not occur without a thorough greyscale examination of the kidneys. Greyscale imaging can provide useful information about renal size and cortical thickness and should be part of the initial series of images. For Doppler image acquisition, a preliminary scan of the abdominal aorta is performed with colour Doppler in the transverse plane beginning at the level of the superior mesenteric artery (SMA) to locate the main renal arteries, which typically originate within 2 cm of the SMA (Fig. 9-1). Transverse images may be obtained from a midline approach with the patient supine or rolled into the left lateral decubitus position. The imager can localise the right renal artery passing posterior to the IVC then rotate the transducer while maintaining the artery in view. Also, the transducer can be placed longitudinally lateral to the rectus muscle resulting in a ‘banana peel’ image (Fig. 9-2), in which the aorta is the banana and the renal artery is the banana skin on each side, being peeled off the banana. If the main renal arteries cannot be demonstrated from the midline approach, a right or left lateral approach is used to follow each artery centrally from the renal hilum. Regardless of the technique used to identify the renal artery origin, the entire main renal artery should be visualised sonographically. Lack of visibility of even a 10 mm segment of main renal artery will limit the sensitivity of the direct method for RAS. This is especially relevant in younger patients in whom fibromuscular dysplasia is a concern; stenosis in these patients may not be near the renal artery origins (described later). As part of the direct Doppler eval‎uation, the peak systolic velocity with angle correction should be measured at the renal artery origin, mid and distal artery, and at any region of turbulent disorganised flow with aliasing on colour Doppler.

FIGURE 9-1 Normal renal arteries. Greyscale transverse image of the aorta demonstrates normal origins of the renal arteries (arrows).

FIGURE 9-2 Normal renal arteries. Longitudinal power Doppler shows the aorta and both renal artery origins in normal location (arrows) with appearance termed as a ‘banana peel’.

Accessory renal arteries occur commonly (approximately 30% of kidneys), but are not always demonstrated sonographically. In fact, studies suggest that only 21–41% of accessory renal arteries are visualised by Doppler eval‎uation.2,3 This low success rate has prompted some individuals to argue that sonographic eval‎uation for RAS is not sensitive enough as a screening study. However, Bude et al. found that less than 1% of accessory renal arteries were the only stenotic artery,⁴ which essentially negates the significance of not visualising an accessory renal artery.

As mentioned earlier, the deep location of main renal arteries often limits their direct eval‎uation, and it will drive the choice of transducer. Lower-frequency transducers will have better sonographic depth penetration, with a trade-off of decreased spatial resolution. As a general rule, the highest-frequency transducer that allows good demonstration of the artery and arterial waveform is preferable.

Greyscale visualisation of the renal arteries should be optimised prior to colour and spectral Doppler eval‎uation. Doppler gain should be adjusted for flow detection by increasing the gain to a level just below the appearance of colour artifact in adjacent structures. Pulse repetition frequency, or velocity scale, is the frequency of sampling, and under-sampling may underestimate peak velocities. In newer systems, built-in software can automatically optimise these parameters, with manual adjustment occasionally required by the sonographer. For spectral Doppler, the Doppler gate should be set to include the entire arterial lumen and angled to the direction of flow. The angle of insonation should be maintained at 60 degrees or less. As angulation increases to 80–90 degrees, the confidence in the measured velocity decreases, as the cosine of the angle of insonation approaches zero. This yields large differences in measured velocity for a small variation in the relative angle of flow.

Segmental Intrarenal Artery Eval‎uation

When the main renal artery is not well seen in its entirety, eval‎uation of the segmental intrarenal arteries may allow a non-diagnostic direct examination to become diagnostic for RAS.5,6 We always examine the segmental arteries even when the main renal arteries are well seen, because the segmental artery waveform morphology may be useful in detecting concomitant renal parenchymal disease.

A posterior flank approach reduces the distance from the transducer to the segmental arteries. Of note, the liver and spleen should not be used as a window to improve visualisation of the kidney, as the vessels are closer to a zero degree angle with the transducer positioned using a more posterior approach. The upper, interpolar, and lower pole segmental arteries are individually studied. A heel–toe technique is commonly applied in which one edge of the transducer is angled into the skin to align the targeted segmental artery flow as close to the transducer angle of insonation (theta less than 20 degrees) as possible to enhance the signal quality. This can enhance the definition of the early systolic peaks. Electronic beam steering can also be used to better align angulation of the insonating beam to enhance waveform morphology.

Characteristics of the spectral Doppler tracing in normal segmental intrarenal arteries should include rapid upstroke to an early systolic peak with gentle decrease in flow velocity during late systole and diastole (Fig. 9-3). Persistent antegrade flow throughout the cardiac cycle should be present without return to baseline. The resistive index (RI), calculated as:

is a common parameter for characterisation of arterial flow. The RI is inversely proportionate to the relative amount of diastolic flow. For instance, an end diastolic velocity that is 20% of the peak flow will result in RI of 0.80. The upper limit of RI in normal adults has been reported as less than 0.70,⁷ but concern for pathology is not often raised until the RI is 0.75–0.80, or higher. Furthermore, the RI may be affected by other factors such as heart rate, Valsalva, and arterial compliance. In fact, RIs greater than 0.70 are common in elderly patients.⁸ The impact of systemic vascular disease in chronic renal dysfunction is significant. It has been recently suggested that the renal RI measurement does not distinguish local from systemic vascular damage. A new potential ultrasound measurement, the difference of RIs between the spleen and kidney, may allow more specific eval‎uation of renal parenchymal damage.⁹ However, this study has not yet been further validated or widely applied in practice.

FIGURE 9-3 Normal segmental artery waveform. Colour Doppler shows normal colour flow without aliasing. Spectral Doppler shows rapid upstroke and smooth velocity decline with persistent flow throughout the cardiac cycle.

Anatomy of the Native Kidneys

ARTERIAL ANATOMY

The renal arteries typically arise from the abdominal aorta caudal to the level of the SMA. The right renal artery usually originates from the anterolateral aspect of the aorta, while the left renal artery usually originates from the posterolateral aspect. As noted earlier, approximately 30% of patients will have more than one renal artery.¹⁰ Accessory renal arteries usually arise from the aorta caudal to the main renal artery to supply the renal lower pole, but occasionally will course cranially to supply the upper pole. Rarely, accessory arteries may arise from an iliac artery or even the SMA. Renal anomalies such as horseshoe or pelvic kidney almost always have multiple renal arteries, which may arise from the aorta or iliac arteries.

The main renal artery divides into dorsal and ventral rami that course posterior and anterior to the renal pelvis. The anterior and superior aspects of the kidney are typically supplied by the larger ventral division. The posterior and inferior portions of the kidney are supplied by the smaller dorsal division. The junction of these ventral and dorsal divisions creates a relatively avascular plane (Brodel’s line), which is the preferred track of percutaneous nephrostomy placement, and should be considered when performing a renal biopsy.

The branching pattern of the renal arteries progresses symmetrically to the renal cortex (Fig. 9-4). Segmental branches arise from the dorsal and ventral rami and run along the infundibulae before dividing into interlobar arteries. These interlobar arteries course between the pyramids, and then branch into arcuate arteries, which run along the bases of medullary pyramids. Within the cortex, small interlobular arteries course outward toward the surface of the kidney.

FIGURE 9-4 Renal arterial branching. Line drawing demonstrates the normal arterial branching from the main artery (M) to the segmental (S), interlobar (IN), arcuate (A), and interlobular (IL) arteries.

VENOUS ANATOMY

The renal venous anatomy parallels the arterial anatomy. Normal venous flow on spectral Doppler has a relatively low velocity. Its waveform is driven by right atrial activity. Accessory left renal veins are less frequent than accessory renal arteries; however, accessory right renal veins are quite common. Left venous anomalies may be seen in approximately 11% of patients.¹¹ Variants most commonly include the retroaortic and circumaortic renal veins (Fig. 9-5), and these may be clinically relevant even beyond filter placement. In a recent study by Karazincir et al., the incidence of retroaortic left renal vein was found to be significantly higher in patients with varicocele, compared with controls¹² (see Fig. 9-24 in the varicocele section).

FIGURE 9-5 Renal vein variants. Line drawings demonstrate the anatomic appearance of a retroaortic (A), and circumaortic (B) left renal vein.

The left renal vein receives drainage from the inferior phrenic, capsular, ureteric, adrenal and gonadal veins and flows across midline into the normal IVC. In patients with a left-sided IVC, the left common iliac vein continues cranially as the left IVC and drains into the inferior aspect of the left renal vein. The right renal vein is shorter than the left and courses obliquely into the IVC. The right renal vein receives capsular and ureteric veins; however, the right inferior phrenic and gonadal veins enter directly into the IVC. Valves may be present within the renal veins, but their reported incidence varies greatly. Renal vein varices may be secondary to renal vein thrombosis or portal hypertension, or they may be idiopathic. Like varicoceles, renal varices are more common on the left than the right. In cases of left renal vein thrombosis, the varices may extend through the inferior phrenic, adrenal, gonadal, and ureteric veins. On the right, the only common branch is the ureteric vein.

Renal Failure and Obstruction

Doppler can play a supportive role in the diagnosis or exclusion of renal obstruction in patients with acute renal failure. Identification of a dilated renal collecting system is fairly easy with ultrasound. The difficulty, however (in the absence of prior examinations) is the differentiation of an acutely obstructed high-pressure system versus that of a low-pressure, chronically dilated system. It has been suggested that elevated resistive index may help differentiate between severe acute urinary obstruction and chronic dilatation.13–15 The RI of the obstructed kidney may be elevated relative to the normal contralateral kidney. An RI difference of greater than 0.10 between the non-obstructed and obstructed kidney is the suggested threshold for diagnosis of acute obstructed uropathy. However, intra-renal autoregulatory hormonal systems counteract the mechanical effect of the high-pressure collecting system pressing upon the parenchyma. This rapidly modifies the resistance to flow, reducing sensitivity of the test. In the setting of partial obstruction or less severe obstruction, this finding also lacks sensitivity.^16,17 As an aside, in cases of chronic renal disease without obstruction, elevated RI > 0.80 has been shown to be associated with worsening renal function and mortality.¹⁸

Another Doppler tool to assist in the eval‎uation of urinary obstruction can be performed within the bladder. In cases with suspected renal obstruction, sonographic eval‎uation for a ureteral jet should be a component of the renal ultrasound examination (Fig. 9-6). Although entry of urine into the urinary bladder is not synchronous, demonstration of three or more ureteral jets by Doppler on one side without a single pulse of flow from the contralateral side implies obstruction of the non-pulsing ureter.

FIGURE 9-6 Normal ureteral jet. Transverse colour Doppler image shows a linear ‘jet’ of colour projecting into the bladder lumen. This represents the rapid flow of urine into the bladder secondary to ureteral peristalsis.