The vast majority of hypertensive patients have essential hypertension unrelated to the renal vasculature. Only a small percentage of hypertensive patients have a renovascular etiology. Renovascular hypertension is more common in patients younger than 20 years of age or older than 50 years of age. In contrast, the onset of essential hypertension is usually between 30 and 50 years of age, and there is usually a family history of hypertension. Rapid acceleration or severe hypertension and severe hypertensive retinopathy also suggest a renovascular etiology. An abdominal bruit on clinical exam of a hypertensive patient should likewise suggest a vascular etiology.

Renovascular hypertension is renin mediated and occurs as a physiologic response to renal ischemia. Renin, an enzyme produced in the juxtaglomerular apparatus of the kidney, acts on the circulating serum protein angiotensinogen to produce the inactive hormone angiotensin I, which is converted to the active hormone angiotensin II. Angiotensin II increases blood pressure by stimulating aldosterone secretion from the adrenal cortex and causes arteriolar vasoconstriction. This process exerts antidiuretic and antinatriuretic effects on the proximal renal tubule by promoting sodium reabsorption.

The afferent arteriole, which acts as a baroreceptor, is the most important factor governing renin release. The transmural pressure across this arteriole may decrease as a result of reduced perfusion pressure or increased compliance of the arteriole. The hypertension in patients with renovascular hypertension is dependent on the high circulating levels of angiotensin II. Therefore, an angiotensin II antagonist or the converting enzyme inhibitor, captopril, may be used to control the hypertension.

Even though many different processes may cause stenosis of the main renal artery, the most common etiologies are atherosclerosis and fibromuscular dysplasia (FMD). Less common etiologies include congenital narrowing, arterial hypoplasia, embolic disease, Takayasu’s arteritis, middle aortic syndrome, irradiation, trauma, or an association with neurofibromatosis. Hypertension may rarely be related to parenchymal abnormalities such as pyelonephritis, glomerulopathies, ureteral obstruction, trauma, and renal mass lesions. Reninoma is a renin-secreting tumor of the juxtaglomerular cells which more frequently occurs in patients under 20 years of age and is more common in females (Dunnick et al. 1983).

Potential clinical criteria used to select patients who should be imaged to evaluate for RAS include the following:

Renal Doppler ultrasound offers distinct advantages as the initial screening tool for RAS. These advantages include lack of ionizing radiation, lack of potentially nephrotoxic intravenous contrast, and lack of potential risk of nephrogenic systemic fibrosis (NSF) in patients with renal insufficiency. NSF is a rare debilitating and occasionally fatal skin disorder, which has recently been associated with intravenous gadolinium contrast administration. In patients with significant renal insufficiency, ultrasound now has become the safest imaging modality for detection of RAS (Broome et al. 2007; Thomsen 2006) with a high rating in the ACR Appropriateness Criteria for evaluation of renovascular hypertension (Brown et al. 2013).

Ultrasound imaging of the renal vasculature can be challenging due to the deep location of the renal vessels, overlying bowel gas, and the large body habitus of many patients. Proper training in the performance and interpretation of the ultrasound examination, as well as a strong-quality control program, are critical to good study quality and diagnostic accuracy. In trained hands, the sensitivity and specificity of renal Doppler ultrasound for RAS are up to 95 and 90 %, respectively (Bokhari and Faxon 2004). Renal Doppler ultrasound technique, however, is an operator-dependent imaging modality and not routinely performed at all clinical programs.

The two main methods for RAS detection are direct visualization of the narrowing and indirect evaluation of the downstream effects of the stenosis on the segmental renal arteries. For direct criteria, it is important to visualize the entire main renal artery at ultrasound. Turbulent flow may suggest RAS. Increased velocities at the point of stenosis may appear as Doppler color aliasing, which should prompt further evaluation with spectral Doppler. A peak systolic velocity (PSV) of greater than 200 cm/s has been suggested as the threshold for Doppler diagnosis of 60 % diameter reduction of the renal artery (Fig. 2) (Olin et al. 1995; Pellerito 2002). Another common criterion is a PSV ratio of the renal artery relative to the aorta greater than 3.5. Although studies have validated both criteria, a recent meta-analysis showed PSV as the best predictor of RAS, with sensitivity and specificity of 85 and 92 %, respectively (Williams et al. 2007). Due to a variety of factors, it may not be technically possible to demonstrate the entire length of the main renal artery. This may create an exam that is indeterminate by direct criteria. In these cases, another tool that can be used is evaluation of the intrarenal arterial waveforms (Stavros et al. 1992). Increased segmental artery acceleration time and loss of early systolic peaks in the segmental arteries should raise suspicion for RAS. Spectral Doppler may show a dampened waveform with spectral broadening and blunting of the systolic upstroke—the tardus-parvus waveform morphology (Fig. 3) (Downey 1998; Kliewer et al. 1993). However, it is important to remember that the absence of the tardus-parvus waveform does not exclude RAS. Vessel compliance may be reduced in patients with atherosclerotic disease, making the tardus-parvus waveform morphology less obvious (Demirpolat et al. 2003).

An additional advantage that renal Doppler ultrasound has over other imaging modalities is its ability to predict which patients may benefit from therapeutic intervention of RAS. In a large prospective study, Radermacher et al. showed that patients with elevated RI of greater than 0.80 will not improve after renal artery stenting (Radermacher et al. 2001). A subsequent study, however, showed that 29 % of patients with renal insufficiency and RI of greater than 0.80 showed improved renal function after revascularization and, perhaps more important, 50 % showed improvement in hypertension (Garcia-Criado et al. 2005).

Continued improvement in multi-detector CT technology has had a profound impact on reconstructed three-dimensional (3D) images that aid in the diagnosis of RAS. Accessory renal arteries are well detected with CTA which is an important advantage of CTA, as many of the false-positive and false-negative results are related to accessory arteries. RAS in either the main or an accessory renal artery are now accurately detected with CTA. The sensitivity and specificity of CTA in detecting RAS of 50 % or more are approximately 90 and 97 %, respectively. The sensitivity is even greater when only stenoses of 75 % or greater are considered (Johnson et al. 1999; Urban et al. 2001). In the absence of renal insufficiency, MRA has the highest rating for evaluation of renovascular hypertension by the ACR Appropriateness Criteria (Brown et al. 2013). MRA techniques used to detect RAS, such as time-of-flight, phase-contrast sequences, and steady-state free precession, may be limited by turbulent blood flow and respiratory motion, but significant improvements have been made in MRA diagnosis of RAS without contrast. 3D gadolinium-enhanced MRA avoids these problems by acquiring a complete data set within a single breath-hold. Contrasted MRA has high sensitivity and specificity for RAS and is better than ultrasound for detection of accessory renal arteries and bilateral disease. CTA and MRA both have the additional advantage of increasing diagnostic accuracy by providing multiple different ways to view the data. These options include multiplanar reconstruction, shaded surface display, and maximum intensity projection techniques.

The role of imaging in detection of renal artery stenosis is now frequently directed by the potential benefit of intervention (O’Neill et al. 2011). Recent studies suggest that renal artery stenting is not always justified. In two recent randomized trials, including the ASTRAL study, there was no benefit with regard to blood pressure or renal function as a result of revascularization of atherosclerotic lesions (Bax et al. 2009; Wheatley et al. 2009).

There is now much controversy in the literature due to the ASTRAL trial results. Recent reviews suggest that there may be a group of hypertensive patients with acute-onset or rapidly accelerating hypertension where parenchymal renal damage has not yet advanced to the level of irreversible hypertension. Evaluation of renal perfusion in cases of renal artery stenosis may assist in determining which patients will benefit from intervention. Recent advances in MR perfusion/diffusion and tissue oxygenation also may offer greater accuracy in predicting potential benefit of intervention.