Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
The tissue contrast and spectroscopic properties of magnetic resonance imaging (MRI) recommend its use as a problem solver for renal and urinary imaging. The unsurpassed ability to discriminate cystic from solid lesions and higher sensitivity to solid, neoplastic elements explains the superiority of MRI compared with other imaging modalities for imaging the kidney. Regarding the collecting system, ureters, and bladder, MRI features exquisite tissue contrast along with robust urographic effects because of the extreme 1) fluid hyperintensity in T2-weighted images and 2) the aramagnetism of excreted gadolinium in delayed postcontrast T1-weighted images. Typical indications for MR imaging of the kidneys, collecting systems, ureters, and bladder include indeterminate renal lesion, postsurgical and postablation follow-up of renal neoplasms, hematuria workup in patients unable to tolerate iodinated contrast, postnephroureterectomy surveillance following urothelial neoplasm resection, bladder carcinoma, pediatric and pregnancy conditions (in order to avoid iodinated contrast and/or ionizing radiation), and urinary obstruction not related to nephrolithiasis.
Technical considerations in MR imaging of the kidneys are essentially the same as for other abdominal indications (see Chapter 1 ) with a few modifications. Subtracted images (precontrast from postcontrast) serve a central role in differentiating benign from malignant renal lesions. Certain common features—such as modest enhancement and precontrast T1 hyperintensity—challenge the human eye to detect or exclude enhancement. Subtractions eliminate precontrast hyperintensity and improve dynamic range to detect subtle enhancement.
The coronal plane is arguably more diagnostically compelling for the kidneys because of its superiority in displaying them bilaterally to assess symmetry. MR urography also favors the coronal plane because of the adaptability to the vertical orientation of the collecting system-ureter-bladder unit and the ability to portray bilaterality ( Fig. 6.1 ). As such, the MR urography protocol differs from the standard abdominal protocol (which typically suffices for nonurographic kidney indications) ( Table 6.1 ). Magnetic resonance urography (MRU) sequences include T1- and T2-weighted varieties. T1-weighted sequences are obtained after contrast, during the excretory phase, usually in the coronal plane as a modification of the dynamic sequence ( Fig. 6.1A ). 2-D and 3-D T2-weighted sequences are performed before contrast administration; otherwise, excreted gadolinium in the collecting systems shortens the T2 of urine, precluding signal in heavily T2-weighted images. The 2-D and 3-D MRU sequences are the magnetic resonance cholangiopancreatography (MRCP) sequences targeted to the collecting systems and ureters; instead of centering on the common bile duct, the slices are oriented to the renal collecting systems and ureters. Lasix administration (5–10 mg intravenously) augments collecting system distention, but presents logistical difficulties, especially in the outpatient setting.
Sequence | Plane | Spatial Encoding | Details |
---|---|---|---|
Steady-state | Coronal or 3-plane | 2-D | T2/T1-weighted; balanced gradients in all axes ➔ insensitive to motion |
Heavily T2-weighted | Coronal | 2-D | Single-shot fast spin-echo technique |
Heavily T2-weighted | Axial | 2-D | Single-shot fast spin-echo technique; cover entire coil sensitivity range |
In-/out-of-phase | Axial | 2-D | Alternatively acquired as Dixon sequence |
T2 MRU | Radial | 2-D | Centered below each kidney; breathhold- or respiratory-triggered with delay between slices |
T2 MRU | Coronal | 3-D | Respiratory-triggered |
T2 MRU | Coronal | 2-D | Repeated eight times with delay between slices (cinegraphic effect) |
Dynamic | Axial | 3-D | Covering kidneys and below |
Postcontrast | Axial | 3-D | Covering pelvis |
T1 MRU | Coronal | 3-D | Covering kidneys, ureters, and bladder; first acquisition flip angle 15 degrees then repeat with 40 degrees |
Diffusion | Axial | 2-D | Covering kidneys through pelvis |
Targeted bladder imaging (usually for bladder cancer evaluation and staging) requires targeted pelvic imaging focusing on T2-weighted imaging primarily and is supplemented by (dynamic) contrast-enhanced pulse sequences. T2-weighted images are performed with high-resolution technique with a time to echo (TE) in the 60 to 100 msec range in the sagittal and axial and/or coronal planes. The utility of dynamic imaging for bladder carcinoma has not been firmly established, but provides complementary information about both the bladder and surrounding structures.
In the context of the kidneys, the first question to consider is whether the abnormality is a focal lesion or a diffuse process. Regarding focal lesions, the chief tissue considerations are: 1) whether a lesion is solid or cystic, 2) whether a solid lesion is a neoplasm or solid nonneoplastic tissue (ie, scar, infarct, or infection), 3) whether a solid lesion contains fat or not, 4) whether a fat-containing lesion contains microscopic/intracellular or macroscopic lipid, and 5) whether hemorrhage connotes an underlying lesion. Discriminating cystic from solid is straightforward and relies on T2-weighted images showing extreme fluid hyperintensity paired with a lack of enhancement. Difficulty in assessing enhancement arises in the case of the T1 hyperintense lesion (such as hemorrhagic cysts). Precontrast hyperintensity plus even more signal (from enhancement) is difficult for the human eye to detect. Subtracted images negate precontrast hyperintensity, showing only changes in signal intensity between the precontrast and the postcontrast image by literally subtracting the signal from each pixel in the precontrast image from each pixel in the postcontrast image. All that remains is a map of contrast enhancement (assuming no intervening patient motion causing misregistration). If subtractions are not available, comparing the region of interest (ROI) with a reference standard establishes or excludes enhancement. As a rule of thumb, normal muscle enhances and serves as a lower limit threshold for renal mass enhancement; most renal tumors enhance avidly, but relatively hypovascular papillary renal cell carcinomas tend to be relatively hypovascular. Compare the ROI signal intensity values of the pre- and postcontrast renal lesion with ROI values of normal muscle (eg, psoas, longissimus). An equal or greater increase compared with muscle indicates enhancement and solid tissue ( Fig. 6.2 ). Diffusion-weighted imaging provides another means of confirming solid tissue and restricted diffusion with lower ADC values compared with cystic lesions. Superimposed septations and mural nodularity or thickening potentially indicates a cystic neoplasm (which is relatively rare). Differentiating nonneoplastic solid lesions, such as scars, infarct, and pyelonephritis is more difficult and often relies on a combination of imaging features, clinical parameters, and longitudinal data (improvement or resolution versus growth).
Identifying and characterizing fat in a solid renal lesion usually confer histopathologic diagnostic certainty and prognostic information. Microscopic fat with out-of-phase (OOP) signal loss generally connotes the clear cell histologic renal cell carcinoma (RCC) subtypes, although a very small fraction of lipid-poor angiomyolipomas (AMLs) shares this imaging pattern. Macroscopic fat appearing uniformly hyperintense and disappearing with fat suppression indicates a benign renal AML. The T1 hyperintensity of hemorrhage occasionally simulates macroscopic fat, but fails to suppress with fat saturation. Whereas RCC often harbors hemorrhage foci, renal or perinephric hemorrhage without an inciting incident (trauma, bleeding aneurysm, arteriovenous malformation, etc.) implies an underlying mass.
Dedicated bladder imaging requires high spatial resolution and bladder distention because the bladder wall is the focus of the examination and underdistention results in wall thickening and redundancy, simulating pathology. For the most part bladder imaging focuses on characterizing and staging bladder carcinoma, which requires an understanding of the normal mural stratification appearance of the pattern of tumor spread.
Normal kidneys measure approximately 10 to 14 cm in length. Renal axes tilt medially at the upper poles with anteromedial rotation (according to the position of the hilum) and the kidneys extend from the T12–L1 to the L3 levels ( Fig. 6.3 ). The kidneys are cloaked in a sheath of retroperitoneal fat with interdigitating fibrous septa capped by the adrenal glands. A barely perceptible linear hypointensity encircles the retroperitoneal fat—Gerota’s fascia—an important landmark in staging RCC ( Box 6.1 ).
T0: no evidence of primary tumor
T1: tumor ≤7 cm limited to kidney
T2: tumor >7 cm limited to kidney
T3: tumor extends into major veins; invades adrenal gland or perinephric tissues not beyond Gerota’s fascia
T3a: tumor invades adrenal gland or perinephric tissues not beyond Gerota’s fascia
T3b: tumor extends into major renal veins or IVC below diaphragm
T3c: tumor extends into major renal veins/IVC above diaphragm
T4: tumor invading beyond Gerota’s fascia
N0: no regional lymph node metastasis
N1: metastasis in a single regional lymph node
N2: metastasis in >1 regional lymph node
M0: no distant metastasis
M1: distant metastasis
Stage I (T1 or 2, N0, M0): tumor confined to kidney
Stage II (T3a, N0, M0): tumor spread to perinephric fat confined with renal fascia; possible ipsilateral adrenal involvement
Stage IIIA (T3b–3c, N0, M0): tumor spread to renal vein, inferior vena cava or both
Stage IIIB (T1–T3a, N1–N3, M0): tumor spread to local lymph nodes
Stage IIIC (T3b–T3c, N1–N3, M0): tumor spread to local vessels and lymph nodes
Stage IVA (T4, any N, M0): tumor spread to adjacent organs (except ipsilateral adrenal gland)
Stage IVB (any T & N, M1): distant metastases
Stage I: T1, N0, M0
Stage II: T2, N0, M0
Stage III: T1–T2, N1, M0 or T3a-c, N0–N1, M0
Stage IV: T4 or any T, N2, M0 or any T, any N, M1
Normal kidneys exhibit corticomedullary differentiation characterized by moderately greater fluid content in the medulla, relative to the cortex ( Fig. 6.4 ). The kidneys enhance avidly with earlier enhancement of the renal cortex during the arterial phase, reiterating the corticomedullary pattern—the renal cortical phase for our purposes. Within 60 to 90 seconds contrast perfuses the medullary renal parenchyma, resulting in global renal parenchymal enhancement—the parenchymal phase.
The renal collecting system is generally biconcave or flat, and excreted urine demonstrates water signal. Occasional flow voids in T2-weighted images sometimes simulate renal calculi. The collecting system urothelial lining appears as inconspicuous linear hypointensity, exhibiting no discernible enhancement.
Congenital anomalies of the kidneys and urinary tract afflict 3 to 6 per 1000 live births, although only a few common developmental anomalies and pseudolesions ( Box 6.2 ) are worth mentioning before discussing pathologic renal lesions. Renal and collecting system embryogenesis is a complex process following a series of steps necessary for the execution of subsequent developmental steps. Incomplete or absent interfacing of the ureteric bud (primordial collecting system and ureter) and metanephric blastema (primordial renal parenchyma) results in renal agenesis and/or a number of other potential parenchymal and ureteral/collecting system anomalies.
Positional anomalies include ectopia and malrotation. Embryonic growth results in a relative ascent of the kidneys during the fourth through eighth week of gestation, ultimately positioned between the first and third lumbar vertebrae. Underascent occurs far more commonly than overascent, and the ptotic kidney ranges in position from the true pelvis to the iliac fossa and anywhere below the expected location centered at the L2 level ( Fig. 6.5 ). Contralateral renal anomalies, such as renal agenesis or ptosis, frequently coexist.
Concomitant medial rotation along the longitudinal renal axis, during ascent, orients the ureteropelvic junction (UPJ) medially. Nonrotation or incomplete rotation leaves the UPJ facing anteriorly, and renal calyces in the medial segment of the kidney lie medial to the renal pelvis ( Fig. 6.6 ). Overrotation results in a posteriorly facing UPJ.
Renal fusion anomalies generally incur positional and rotational derangements. Medial renal fusion results in a solitary discoid lump of renal tissue in the pelvis, referred to as “pancake kidney.” Crossed fused renal ectopia represents the sequela of embryologic renal fusion with the relatively normally positioned kidney dragging its fused counterpart across the midline, resulting in a single ipsilateral S-shaped renal mass with two separate moieties and normal bilateral ureterovesical junctional positioning ( Fig. 6.7 ). Horseshoe kidney is the most common renal anomaly reflecting midline fusion of the metanephric blastema. Ascent is arrested at the level of the inferior mesenteric artery ( Fig. 6.8 ). Coexistent anomalies, such as UPJ obstruction and duplication anomalies, conspire with the geometric and rotational distortion and urinary stasis to lead to complications, including stone formation and infection.
Structural anomalies incur no risk of complications and deserve mention only to prevent misdiagnosis. Fetal lobulation persists in 5% of adults with a smooth undulating outer renal contour conforming to the position of the renal pyramids. Smoothly marginated indentations conform to the edges of renal pyramids with normal appearance and thickness of underlying parenchyma (≥14 mm), excluding the possibility of an underlying mass or scarring. The column of Bertin potentially simulates a renal mass, but represents invagination of normal renal cortical tissue into the renal sinus, usually occurring at the upper polar/interpolar junction and averaging 3.5 cm in size ( Fig. 6.9 ). The hilar lip represents fusion of medial renal lobes, usually occurring in the upper pole and potentially protruding and distorting the renal sinus. Signal characteristics and enhancement identical to the adjacent renal parenchyma, occurring in the expected location, confirm the presence of normal functional renal tissue in cases of renal anomalies.
Focal renal lesions are encountered every day in clinical practice in cross-sectional imaging studies—the vast majority of which are incidental, and many indeterminate. Thirteen percent to 27% of abdominal imaging studies incidentally detect a renal lesion. Whereas many of these lesions are incompletely characterized, the overwhelming majority of these lesions are simple or minimally complicated cysts with no malignant potential. Establishing true cystic etiology eliminates the need for further workup and/or follow-up. The presence of solid components implies malignancy, which usually mandates surgical resection ( Fig. 6.10 ).
Most renal cysts are simple in appearance with fluid signal characteristics (T2 hyperintensity and T1 hypointensity), no enhancement, septation, wall-thickening, or nodularity. However, simple cysts occasionally experience complications in the form of hemorrhage, infection, rupture, etc. As such, these benign, nonneoplastic cysts are referred to as “complicated cysts,” to distinguish them from complex/neoplastic cysts, which demonstrate some element of solid tissue. Signal alterations alone pose no risk of malignancy, although they often challenge interpretation. The most common signal alteration occurs as a result of hemorrhage, resulting in T1 hyperintensity and T2 hypointensity. Proteinaceous contents induce a similar appearance.
Other complicated cystic lesion features pose greater diagnostic difficulty. Superimposed infection and trauma induce reactive wall thickening, which overlaps with the appearance of cystic neoplasms (especially RCC, clear cell type). These neoplasms often harbor more complex features with mural nodularity and solid components. In an effort to stratify renal lesions based on the likelihood of malignancy, Bosniak developed a predictive classification system (for computed tomography [CT]) to guide management ( Table 6.2 ). Although not specifically adapted for MRI, this scheme illustrates the range of imaging complexity of renal lesions and offers management guidance. Although the MR inconspicuity of calcification precludes factoring it into the classification scheme, it generally applies to magnetic resonance (MR) findings—replacing intensity changes for density changes.
Bosniak Category | Imaging Features | Management |
---|---|---|
I | Thin wall; no septa, calcifications, or solid components, water features; no enhancement | No follow-up |
II | Few thin septa with less than or minimal enhancement; fine calcification or focal thick calcification; homogeneous high-density sharply marginated lesion (<3 cm) with no enhancement | No follow-up |
IIF | Multiple thin septa with less than or minimal enhancement; thick or nodular calcification; no enhancing soft tissue components; high-density lesions (>3 cm) with no enhancement | Observe |
III | Thickened irregular or smooth walls or septa with enhancement | Surgery |
IV | Same features as III with enhancing soft tissue components | Surgery |
Solid lesions include benign and malignant neoplasms. Macroscopic fat is the only finding that connotes a benign etiology—AML. After excluding AML, renal infection and infarction, rare benign neoplasms (such as oncocytoma), and solid tissue (enhancement) effectively equals malignancy. Unless clinical findings raise the suspicion of inflammatory or vascular etiology, the presumptive diagnosis is malignancy until proved otherwise. Because biopsy results are notoriously confusing, solid renal lesion management involves either percutaneous ablation or surgical pathologic diagnosis and treatment. Because very small lesions (<1 cm) challenge the resolution of imaging methods, limiting diagnostic confidence, imaging surveillance preempts potentially unnecessary surgery ( Table 6.3 ).
Size | Presumptive Diagnosis | Management |
---|---|---|
Large (>3 cm) | Renal cell carcinoma | (Partial) Nephrectomy ∗ |
Small (1–3 cm) | Renal cell carcinoma | (Partial) Nephrectomy ∗ |
Very small (<1 cm) | Renal cell carcinoma, oncocytoma, angiomyolipoma | Active surveillance |
∗ Ablative treatment is an option in the appropriate clinical setting.
Using the combination of heavily T2-weighted images to detect fluid hyperintensity in postcontrast images to exclude enhancement confirms cystic etiology. T1-weighted sequences depict hemorrhagic cysts with variable hyperintensity, and postcontrast images exclude enhancement. Ninety percent of all renal cystic lesions are simple cysts ( Fig. 6.11 ). Hemorrhage, debris, and infection complicate renal cysts. If there is an increasing number of cystic lesions (and renal enlargement), consider the possibility of polycystic disease. Cystic lesions prevail in other inherited diseases, such as von Hippel–Lindau (VHL) disease and tuberous sclerosis (TS) (with assorted solid lesions, discussed in the section “Solid lesions with ball morphology”). Developmental etiologies include multicystic dysplastic kidney and calyceal diverticulum. Acquired conditions, such as renal cystic disease of dialysis and lithium therapy, present with renal cystic lesions. RCC (clear cell types) dominates the cystic neoplastic category; consider multilocular cystic nephroma (MLCN) in the appropriate demographic categories (young males and middle-aged females), with herniation into the renal pelvis.
The simple renal cyst is ubiquitous, with a prevalence of up to two thirds of the population, and increasingly prevalent with age. Renal cysts detach from the parent renal tubule and become self-enclosed; continued fluid secretion distends the cavity, resulting in an isolated cystic structure. Ongoing fluid secretion accounts for continued growth of simple renal cysts (≤5% per year), despite the lack of neoplastic or autonomously regenerating cells. Therefore careful attention to imaging features is paramount in excluding neoplasm.
The contents of the simple renal cyst—free water protons—account for its appearance: extreme T2 hyperintensity and T1 hypointensity with no enhancement. If even visible, the cyst wall is uniformly smooth with no perceptible enhancement in postcontrast images ( Fig. 6.12 ). Size varies from millimeters to over 10 cm. Cysts are stratified into three different categories based on their location: 1) exophytic, 2) parenchymal, and 3) parapelvic ( Figs. 6.13 and 6.14 ). Because most renal cysts arise from the cortex (although some develop from the medulla), most renal cysts are exophytic and/or intraparenchymal (see Figs. 6.12 and 6.13 ).
Renal cyst complexity assumes many forms, which fall into two major categories—morphologic and signal-related. Morphologic derangements include deviation from simple sphericity and/or septation; signal derangements deviate from simple fluid characteristics with evidence of hemorrhage or debris ( Fig. 6.15 ). Occasional fluid-fluid levels depict layering hematocrit ( Fig. 6.16 )—a relatively infrequent manifestation of a hemorrhagic cyst. Signal alterations from hemorrhage present the greatest challenge to interpretation; precontrast hyperintensity limits the ability to detect augmented T1 signal from enhancement (see Fig. 6.15 ). Fibrinous septations from previous infection generally measure less than 2 mm in thickness with no other evidence of complexity to suggest neoplasm ( Fig. 6.17 ). As long as no enhancement beyond minimal linear septal or wall enhancement is present, simple cystic etiology is confirmed.
Careful scrutiny of these features is critical, because a minority of clear cell RCC (less than 10%) is mostly cystic. In Bosniak’s terms, the discrimination of types II and IIF lesions from understated type III lesions bears close inspection. Usually, a solid enhancing component differentiates RCC from a nonneoplastic cyst—especially with the benefit of subtractions. When subtractions are not available and signal alterations limit assessment of enhancement, rely on ROI measurements compared with a reference standard (see Figs. 6.2 and 6.15 ). Other etiologies are not realistic considerations, except under specific circumstances. For instance, with multilocularity and when herniating into the renal pelvis, consider multilocular cystic nephroma (MLCN) in the appropriate demographic setting (young males or middle-aged females). Infectious cysts—renal abscesses and echinococcal cysts—require a suggestive history.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here