Renal ultrasound, which is fast, noninvasive, and easily accessible, is widely used in the initial evaluation of the urinary system (Figure 2A). Renal ultrasound can be used to assess the size, shape, and location of the kidneys, and look for the presence of hydronephrosis, a sensitive-but-not-specific sign of UTO. As with other anatomic imaging modalities, renal ultrasound can identify the presence of hydronephrosis but cannot generally determine if that hydronephrosis is due to a UTO or a non-obstructive pathology. Increases in renal cortical echogenicity, assessed relative to the liver if the right kidney is in situ, are generally indicative of underlying medial renal disease. Furthermore, in most instances, renal masses and stones are also evident. Nonetheless, definitively identifying the etiology of a UTO based on ultrasound is often challenging because of several inherent limitations [5,6]: (1) acquisition and interpretation of ultrasound images depends on the operator; (2) renal ultrasound cannot visualize the full length of the ureter unless it is significantly dilated; (3) the resolution and tissue penetration depth are not sufficient to detect smaller pathological defects. Even with these drawbacks, ultrasound is still used as a valuable first-line investigation to rule out obstructive nephropathy in patients presenting with laboratory evidence of acute kidney injury (AKI) [7] and/or suggestive symptoms. Additionally, contrast-enhanced ultrasound (CEUS) using microbubbles is an emerging technique increasingly employed for imaging of the kidneys. At this stage, CEUS is primarily used for the evaluation of renal masses, although it has also shown promise in assessing renal perfusion.

Intravenous and retrograde urography have been mainstays for renal imaging prior to the advent of cross-sectional imaging. In X-ray urography the urinary system is opacified by iodinated contrast, which is administered intravenously and excreted or retrograde via foley catheter [8]. While reliably detecting hydronephrosis (Figure 1B), X-ray urography provides limited sensitivity in detecting many important urological pathologies, most notably small urothelial lesions and renal masses. Furthermore, while IVU may demonstrate delayed excretion, indicating functional impairment, improved diagnostic accuracy can be achieved with newer imaging methods, including CT and MRI.

Computed Tomography (CT), which utilizes a helical array of X-rays to produce volumetric tomographic images, has largely supplanted IVU in the assessment of UTOs (Figure 2B). CT provides excellent spatial and contrast resolution and utilizes iodinated agent contrast, effectively delineating the kidneys, collecting systems, ureters, and bladder [9,10,11,12]. Unenhanced CT, while lacking the contrast resolution of a multi-phasic contrast enhanced exam, is nonetheless effective in identifying hydronephrosis, and is the standard bearer for detection of radiodense stones. The presence of hydronephrosis, without a discernable cause, is often further assessed with multi-phasic CT. By obtaining non-contrast, corticomedullary (approximately 30 s), nephrographic (60 s), and delayed phases (5–8 min), one can identify otherwise occult causes of UTO, most notably renal/urothelial tumors. Furthermore, the persistence of nephrographic appearance in the hydronephrotic kidney and absent or reduced contrast excretion indicates accompanying functional impairment, although again the degree of impairment cannot be reliably quantified. Further, hydronephrosis on CT can indicate either UTO or non-obstructive pathology, and in many cases the distinction cannot be determined based only on imaging findings. Radiation exposure is an important limitation of CT, as well as contrast-induced nephropathy (CIN) by iodine contrast media, which present risks for patients with impaired kidney function. Despite these limitations, CT remains a mainstay of renal imaging.

Nuclear scintigraphy is currently the gold standard for the assessment of differential kidney function and for evaluating UTOs. The fundamental reason for the predominance of nuclear medicine scintigraphy in evaluating UTOs is the ability to differentiate a patulous renal collecting system from an obstructive uropathy [13]. This approach is based on time resolved planar imaging of the kidneys after intravenous administration of a radiotracer that is either filtered (^99mTc-diethylene triamine pentacetic acid, ^99mTc-DTPA) or secreted by the tubules (^99mTc-mercapto acetyl triglycine, ^99mTc-MAG3). As such, nuclear medicine scintigraphy is fundamentally based on physiology and not anatomy. In cases in which a patient has evidence of hydronephrosis from anatomic imaging, the patient should undergo nuclear medicine scintigraphy in order to determine whether the hydronephrosis is due to a UTO or a result of non-obstructive pathology. In fact, ^99mTc-MAG3 is particularly useful due to its mechanism of clearance being tubular secretion-based; as a result, it is usable even with depressed eGFR due to renal dysfunction [13].

Scintigraphic evaluation can be enhanced by use of Lasix as a diuretic. In situations in which the clearance of radiotracer from the renal collecting systems is slow or absent, Lasix can be administered in order to increase the production of urine and overcome any intrinsic patulousness in the system. If a hydronephrotic collecting system clears rapidly after administration of Lasix, the system is dilated but not obstructed. If, on the other hand, the system clears slowly or not at all (Figure 3), the findings are indicative of functional obstruction. This use of physiology to discern the nature of hydronephrosis is unique among currently available imaging techniques. However, this technique suffers from low spatial resolution and does not provide detailed anatomic information; therefore, a cause of UTO may not be appreciated. Renal scintigraphy is sensitive to patient details and does involve ionizing radiation—therefore, it can have limitations in use in pediatrics [13].

MR urography is an alternative technique for the detection of urothelial lesions and strictures. MRI utilizes magnetic fields and radiofrequency pulses without imparting ionizing radiation and displays soft tissue contrast due to differences in water density and relaxation. Two MR urography methods are currently employed: static-fluid MR urography and excretory Dynamic Contrast Enhanced MR urography (DCE-MRU). Static-fluid MR urography exploits the long transverse relaxation times (T₂) of urine, with heavily-fluid weighted sequences that suppress short T₂ proton signals with long echo times, typically acquired using single-shot fast spin-echo (SS-FSE) and half-Fourier single-shot spin echo (HASTE) sequences to provide outstanding visualization of the urinary tract. This is of particular concern in UTO, where an obstructed kidney may generate little urine. Furthermore, dynamic cinematic acquisitions can be obtained, allowing areas of ureteral peristalsis to be distinguished from fixed stenoses. Excretory DCE-MRU urography currently relies on excreted MRI imaging agents (mainly based on gadolinium) to generate signal changes within the collecting system (fat suppressed T₁ acquisition, typically obtained using gradient-echo techniques). Both MRU approaches avoid use of ionizing radiation and pose no risk of CIN, which are advantages over CT. They have none of the technical limitations of US. Static fluid sequences do not require administering gadolinium contrast, avoiding concerns related to NSF and cumulative gadolinium deposition. As shown in Figure 2C and Figure 4, the resulting images possess better soft tissue contrast compared to other modalities. In practice, static and DCE-MRU are often employed in tandem with accompanying multiphasic imaging of the kidneys and/or bladder.

CEST has emerged as an MRI contrast mechanism that can be used to detect small amounts of contrast agent through an amplified detection of low concentration protons (on the order of low mM) through their exchange with water [38]. Based on the early work of Cerdan, Gillies, Sherry, Bhujwalla and colleagues, pH imaging is an emerging field [39,40,41,42,43,44,45,46]. This includes the outstanding work of Andreev, Reshetnyak, Lewis, Gao and colleagues developing both fluorescent and PET probes to detect low pH environments [47,48,49,50,51]. CEST MRI is now one of the premier imaging methods for the measurement of pH due to proton chemical exchange being sensitive to acid-base equilibrium, the development of ratiometric methods for distinguishing changes in agent concentration from changes in pH, and the amplified detection of these agents compared to spectroscopy [52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67]. Traditional CT and MR urography, while effective at characterizing upper tract function, cannot reliably quantify renal function, often requiring subsequent renal scintigraphy exams. Conversely, CEST MRI can characterize urinary tract obstruction and function simultaneously, generating pH maps in addition to traditional time activity curves seen on renal scintigraphy. The novel ability of CEST imaging to measure pH is of particular import, as renal functional impairments are often associated with a urinary acidification defect caused by diminished net H⁺ secretion and/or HCO₃⁻ reabsorption. While this can be assessed systemically through measurement of urine pH, using MR to image tubular pH can best assess whether a patient is experiencing functional decline in their obstructed renal unit, quantify the amount of renal functional impairment, and even assess whether this functional impairment is reversible (thus potentially benefiting from intervention) or irreversible. Additionally, in the case of urolithiasis, pH is relevant due to its relationship with stone formation, with urine acidification or alkalization employed as treatments depending on the stones’ constitution. Hence, CEST imaging is of interest in the evaluations of upper UTOs not only due to efficiency, but the ability to provide novel functional information to inform clinical therapy.

The triiodobenzenes iopamidol (ISOVUE^®, Figure 5A) and iopromide were the first agents with particular promise identified for CEST imaging of the kidneys [66,68]. These are safe, nonionic molecules that have been in clinical use for over 30 years as X-ray contrast agents and administered at very high doses (up to 400 mg/mL). They both also have two exchangeable amide proton resonances (for iopamidol 4.2 ppm and at 5.5 ppm downfield from the water signal, Figure 5B) that produce pH-dependent CEST contrast. Ratiometric methods for pH assessment have been developed based on comparison of the signals at the frequencies of these two amide proton resonances to measure pH within the range of 5.5 to 7.4. Longo and colleagues evaluated if iopamidol-based CEST MR could be used to detect the recovery of kidney function in an ischemia reperfusion acute kidney injury (AKI) rodent model using two metrics: renal perfusion and renal pH [69]. A return to normal perfusion and pH values was observed by Day 7 for moderate lengths of ischemia induction, whereas a persistent drop in the perfusion of contrast agent and increase in renal pH was observed for longer induction times (Figure 5C,D). Our group evaluated if iopamidol-based CEST MR could be used to detect progression in chronic kidney disease (CKD) in a the methyl malonic acidemia (MMA) model [70]. This was investigated using four groups of mice: healthy controls on a regular diet (RG Mu^+/−) and high protein diet (HP Mu^+/−), mild kidney disease mice kept on a regular diet (RG Mu^−/−), and severe kidney disease mice on a high protein diet (HP Mu^−/−). Both RD and HP Mu^+/− controls displayed homogeneous pH values of 6.5 and excellent kidney perfusion of agent (~100%) while, in contrast, RD Mu^−/− mice displayed a lower average pH (~6.1) and perfusion (~79%) and an order of magnitude larger range of pH values in the kidneys (±0.2). Furthermore, HP Mu^−/− mice displayed a slightly lower pH (~6.0), substantially lower perfusion (~15%) and significantly larger range of pH values (±0.45). We have further tested iopamidol-based CEST MR on a complete unilateral urinary obstruction (UUO) mouse model. As shown in Figure 6, there are, again, changes in both iopamidol perfusion, average pH and range in pH values could be visualized. Other important work has been performed to improve the imaging protocols and establish these agents on different models of kidney disease and for tumor imaging as well [37,56,71,72]. The results have demonstrated that iopamidol-based CEST MRI pH mapping is promising for monitoring of renal function, allowing for an early detection of the occurrence of renal pathology and distinguishing moderate from severe AKI.

Other agents have also shown promise for kidney imaging. For example, Sherry and colleagues performed in vivo pH imaging of mouse kidneys using a frequency-dependent lanthanide-based CEST agent [73]. Other agents tested include urea and phosphocreatine [74,75]. Our group has synthesized imidazole CEST imaging agents including Imidazole-4,5-dicarboxamide-di Glutamate (I45DC-diGlu) [76], and, as is shown in Figure 7, seems particularly promising for evaluating UTOs. This agent is well suited for ratiometric pH imaging on 3 Tesla scanners with two labile protons with large chemical shifts that produce strong pH sensitive contrast (Figure 7A,B). Figure 7C,D display higher kidney contrast and a larger difference in split renal contrast than seen using iopamidol. A number of well-tolerated medications have been developed based around imidazoles [77], including for treatment of cancer [78], ulcers [79,80], hypertension [81] and as antihistaminic drugs [82] which is encouraging for translating imidazole MRI contrast agents. We expect that one of the CEST agents mentioned above, or perhaps one not yet discovered, will prove outstanding for functional DCE-MR CEST urography of UTOs.

At this stage, all of the major clinical scanner vendors have sequences for performing CEST imaging, allowing for the translation of pH imaging to patients [83,84,85,86,87]. Based on the success of our studies in mice, we prepared an iopamidol phantom at several pH values in serum and moved to establishing a CEST imaging protocol on our 3T Philips Achieva scanner (Philips Healthcare Solutions, Amsterdam, NL, USA) to test how well we could create pH maps. Example data are shown in Figure 8. As is shown, pH mapping could be performed over the range of 5.9 to 7.3 on our scanner; however, below this pH the performance was not ideal. Based on this, we injected iopamidol into our first healthy subject (Isovue 300, 90 mL injection volume) and observed ~4% contrast across both kidneys which was strong enough to enable generation of our first pH maps on a human subject (Figure 8D). Improvements in imaging protocol and post processing are currently being implemented and are expected to yield higher contrast to noise ratio pH maps. These results indicated that DCE-MR CEST urography is very promising for translation into patients with UTOs.