Marshall L. Stoller, Maxwell V. Meng-Urinary Stone Disease

Подождите немного. Документ загружается.

380 Breiman and Coakley

Fig. 11. Perinephric fat stranding (arrow) around the left kidney caused by a distal obstructing

stone (not shown). Perinephric fat stranding in this setting reflects edema and distended lymphat-

ics, and is often associated with a loss of definition of renal contours.

Fig. 12. (A) Contrast enhanced CT in a patient with metastatic breast cancer and acute left sided

flank pain. Images at a lower level (not shown) showed a 5 mm stone in the left ureterovesical

junction. Note the presence of fluid (arrow) around the left kidney. (B) CT image obtained 15

mins later. Excreted contrast is seen to pass from the left pelvicaliceal system into the left

perirenal space (arrow), confirming the diagnosis of forniceal rupture secondary to calculous

obstruction.

Chapter 20 / Imaging of Urinary Stone Disease 381

peristalsis, with a potential paradoxical decrease in pain. Complete obstruction lasting

4–7 d or partially for 14 d may result in various degrees of irreversible decrease in renal

function (24,25).

SPECIFIC IMAGING MODALITIES IN URINARY STONE DISEASE

Overview: Imaging Evaluation of Acute Flank Pain

The primary diagnostic consideration in a patient with acute flank pain is an obstruct-

ing ureteral stone. Consequently, studies investigating imaging algorithms typically use

identification of a ureteral stone as the end-point. This obscures the fact that the radio-

logical evaluation of acute flank pain should answer two related but distinct questions—

is urinary obstruction present and, if so, what is the level and cause of obstruction? The

presence or absence of ureteral obstruction is a functional question, whereas cause and

level are anatomic issues. Many imaging tests are primarily anatomic or functional in

nature. The apparently conflicting views on appropriate imaging in acute flank pain

often reflect a conceptual failure to distinguish these two issues. US and noncontrast

enhanced CT (NCCT) use the anatomic findings of hydronephrosis and hydroureter as

indirect indicators of obstruction, but do not provide direct functional assessment.

Only contrast-enhanced CT and IVU provide direct functional information. Plain radio-

graphs provide no meaningful evaluation of obstruction, but can directly demonstrate

the cause and level of obstruction, if obstruction is a result of a radiographically demon-

strable calculus. US rarely directly visualizes the cause of obstruction. CT can depict the

cause and level of obstruction in nearly all cases.

Plain Radiography

In the past the first imaging study requested in patients with acute flank pain was

frequently a plain radiograph (also known as a conventional abdominal film, abdominal

flat plate, or KUB; the latter refers to inclusion of the kidneys, ureters, and bladder in the

field of view). Detectability of stones on conventional radiographs depends on size and

composition as well as factors relating to technique, patient body habitus and overlying

structures, such as bowel contents. In theory 90–95% of stones are sufficiently radio-

paque to be visible by plain radiography (26). In practice, the plain radiograph is limited

in the evaluation of renal colic because multiple radiodensities can mimic stones (e.g.,

gallstones, costochondral calcifications, bone islands, and fecal densities) and stones

may be easily missed (e.g., radiolucent stones, and stones obscured by bowel content or

bone) (Fig. 13). Bowel preparation has been traditionally thought to improve detection

of stones, but a recent randomized study failed to confirm this hypothesis (27). These

limitations are illustrated by the reported accuracy of plain radiographs for renal colic.

In a study of 51 patients, using IVU as the gold standard, the sensitivity was 29% and

specificity was 73% (28). In a study of 49 patients, using stone passage or retrieval as

the gold standard, plain radiography had a sensitivity of 68% and a specificity of 96%

(29). In another study of 40 patients, using stone passage or retrieval as the gold standard,

plain radiography had a sensitivity of 54% and a specificity of 67% (30). The sensitivity

of a plain radiograph for the detection of urinary calculi ranges from 45 to 58% when

NCCT is used as the standard (31–36). The addition of nephrotomography improves

radiographic sensitivity by 30–40%, but patient radiation dosage is increased signifi-

cantly and overall accuracy remains well below NCCT (35). Plain radiographs remain

useful in the planning and the guidance of shock-wave lithotripsy (SWL) and in moni-

382 Breiman and Coakley

toring the progress of a stone treated conservatively or stone fragments after lithotripsy,

provided the stone can be seen on plain radiographs and has been confirmed by CT.

In some centers, digital radiographs (DR) or computed radiographs (CR), and scanned

projection radiographs (SPR), including CT scout radiographs, serve as an alternative to

conventional film screen-based radiographs for stone detection and follow-up. The scout

radiograph, obtained at the time of a diagnostic CT, may be used as a baseline study in

Fig. 13. (A) Plain abdominal radiograph in a patient with left flank pain. A small calcific density

(arrow) is seen projected over the left sacroiliac joint, and was initially interpreted as suggestive

of a ureteral calculus. (B) Axial CT image (shown at bone windows) through the level of the

density seen in (A) shows that the appearance is caused by a bone island (arrow) in the left side

of the sacrum. This example illustrates the known limitations of plain radiography in the assess-

ment of flank pain, because there are multiple causes of both false positive and false negative

results.

Chapter 20 / Imaging of Urinary Stone Disease 383

those patients to be followed with plain radiographs and may help in predicting the

success of radiographs for monitoring the progress of stone passage and whether stone

opacity is adequate to allow radiographic or fluoroscopic guidance of lithotripsy plan-

ning (32). Digital scout radiographs may also help predict stone composition, as a large

stone seen on CT but not evident on the scout view likely contains little calcium and a

high concentration of uric acid or xanthine (32). The resolution of digital radiographs is

slightly less than that of conventional radiography. Averch (37) showed the accuracy of

these modalities to be similar for detection of stones, whereas Assi (31) demonstrated

a significantly lower sensitivity of the scout radiograph compared with conventional

radiography, particularly for stones 3 mm or less in diameter.

Intravenous Urography

Until the advent of spiral CT, the excretory or intravenous pyelogram or IVU was the

gold standard for the evaluation of urolithiasis for over 70 yr (Fig. 14). It is a relatively

safe and accurate means of diagnosing and characterizing urinary calculi, yielding infor-

mation on renal function as well as anatomy. The IVU is particularly useful in document-

ing complex or subtle abnormalities of the urinary tract that may lead to stone formation

or complicate management. Nonopaque radiolucent stones can be detected as filling

defects within the collecting system, ureters or bladder. An IVU can usually differentiate

renal calcifications and calculi from extrarenal calcifications but may not always be able

to differentiate renal parenchymal calcifications from calculi. The location and size of

a calculus, the precise relationship of a renal calculus to the renal collecting system, and

the degree of resultant obstruction can be assessed with an IVU. Underlying anatomic

abnormalities predisposing to stone formation, or affecting the choice of therapy, can be

assessed as well. A logistical disadvantage of IVU is that a physician must be present

because of the risk of contrast reactions. This may limit the availability of IVU in small

departments with fewer personnel or off-site coverage. Another disadvantage is that the

study may be protracted in patients with obstruction, when delayed images may be

required several hours after injection of contrast for complete assessment. The emer-

gence of NCCT as the primary test in patients with flank pain also exposes physicians

who still request or perform IVU for this indication to medicolegal liability in the event

of a contrast reaction. In a case where an emergency IVU resulted in a fatal contrast

reaction, the subsequent settlement distributed 90% of the responsibility to the perform-

ing radiologist and 10% to the ordering urologist, despite an attempt by the radiologist

to convince the urologist that a noncontrast CT had become the community standard.

Criminal manslaughter charges were also pressed, but were eventually dropped (38).

Although the IVU for flank pain is rapidly passing into history, the imaging findings

of urolithiasis on IVU remain of interest. Following calculous ureteral obstruction, the

nephrogram may initially be normal, but later becomes dense and associated with

delayed excretion as tubular filtration decreases (Fig. 14). Fine striations related to

dilated tubules may be present in the area of the renal pyramids. Opacification of the

obstructed collecting system may take as long as 24 h when the obstruction is severe.

As excreted contrast is heavier than unopacified urine, a change in patient position may

be helpful in opacifying previously unopacified portions of the collecting systems or

ureters. Radiolucent stones may be identified as filling defects on an IVU (Fig. 4). If

a stone is similar in density to excreted contrast, it may not be recognized on excretory

films, and may be only visible on the precontrast scout radiograph. An IVU may be

helpful in differentiating intra- from extraluminal opacities. A stone impacted at the

384 Breiman and Coakley

ureterovesical junction may create an apparent bladder wall mass on an IVU. This

pseudomass is most frequently associated with stones lodged in the ureterovesical

junction. Edema surrounding an impacted UVJ stone may create a pseudoureterocele.

Nodular bladder wall edema may simulate a bladder neoplasm. Overall, although IVU

Fig. 14. (A) Scout radiograph from an IVP shows a 2 cm calcific density (arrow) at the left L3/

4 disc space. (B) Five-minute film shows asymmetry of the upper tracts. The right nephrogram

and pelvicaliceal system are unremarkable, whereas the left kidney is enlarged with a dense

nephrogram and delayed excretion. These findings are typical of obstruction. (C) Thirty-minute

film shows contrast excretion into a dilated left pelvicaliceal system (arrow) and ureter, to a level

of the previously noted calcific density. The findings confirm the diagnosis of an obstructing

ureteral calculus.

Chapter 20 / Imaging of Urinary Stone Disease 385

reliably demonstrates obstruction, the cause is not always clearly delineated. In a series

of 46 patients with calculous ureteral obstruction, the stone was not visible at IVP in 6

(13%) (39). The six missed stones were confirmed by retrograde pyelography, ureter-

oscopy, or spontaneous passage. In two other series, obstructing stones were missed by

IVU in 12 of 28 and 6 of 11 patients (30,40). The emergence of NCCT probably

accounts for the apparent drop in IVU sensitivity for stone identification. In the past,

an IVU showing mild hydroureteronephorosis but no stone was often reported as “pos-

sible recent stone passage.” It is possible that many of these cases were caused by small,

urographically occult stones.

Persistent urinary obstruction may result in forniceal rupture. The resultant decrease

in pressure from the forniceal rupture may result in symptomatic relief and be mistaken

for stone passage. A forniceal rupture of short duration and in the absence of infected

urine is usually of no clinical significance. Perinephric fluid resulting from a forniceal

rupture is important to differentiate from perinephric edema and fat stranding in a patient

with infected urine, however, as prophylactic antibiotic treatment may be administered

to avoid perinephric abscess formation. Other findings of acute ureteral obstruction

include an enlarged ipsilateral kidney (nephromegaly), and hydronephrosis or hydro-

ureter.

Intravenous iodinated contrast may be required for IVU or CT urography (CTU), so

a brief discussion of safety is warranted. Contrast reactions range from mild hives with

pruritis to bronchospasm, hypotension and anaphylactic reaction or death. Low osmo-

lality nonionic contrast agents, available since the late 1980s, have reduced the risk of

severe reactions by approx 80%. The risk of fatal anaphylaxis following intravenous

contrast is approx 0.9/100,000 with both high and low osmolality contrast (41). Premedi-

cation regimens including steroids and both histamine-1 and histamine-2 blockers re-

duce the risk of contrast reactions, but have not been shown to specifically reduce the risk

of death. Nephrotoxicity is the other major adverse effect of intravenous contrast. Strat-

egies to reduce the frequency of nephrotoxicity, particularly in high-risk patients such

as those with diabetes or renal impairment, include hydration, premedication with

acetylcystine, or use of iso-osmolar dimeric contrast media (41–43).

Ultrasound

US may be used to directly depict stones, to identify hydronephrosis in calculous

ureteral obstruction, or to guide therapy such as percutaneous nephrostomy tube place-

ment, SWL, and percutaneous nephrolithotomy. The advantages of US include safety

(no iodinated contrast or ionizing radiation), availability, and lower cost relative to

CT. The disadvantages are related to limited accuracy, and will be discussed in greater

detail.

Calculi reflect sound creating echogenic, bright foci with acoustic shadowing deep to

the calculus (Fig. 15). Calculi that have a high percentage of matrix rather than calcium

may be echogenic without acoustic shadowing. Small stones, less than 5 mm, may be

difficult for US to confidently diagnose, as they are less likely to cast an acoustic shadow.

The use of a higher frequency transducer, 5–10 MHz, increases the echogenicity of small

stones and accentuates shadowing, but is associated with decreased beam penetration,

which often limits its application particularly in medium sized to large patients. Renal

parenchymal calcifications and focal echogenic renal sinus fat may be difficult to dis-

tinguish from stones by US. The difficulty in depicting small echogenic stones against

a background of echogenic renal sinus fat should not be underestimated, and represents

386 Breiman and Coakley

a major limitation of sonography for diagnosis of renal calculi. In a recent study using

nonenhanced CT as the gold standard, US only detected 24 of 101 calculi, with a speci-

ficity of 90% (44).

US is used to identify obstruction by demonstration of hydronephrosis rather than by

direct visualization of a stone (although an obstructing calculus may sometimes be seen

high in the ureter through the acoustic window for the kidney, or in the distal ureter

through the bladder). Bowel and bone limit acoustic access to the rest of the ureter in

the retroperitoneum. The main drawback of US is that not all obstructed kidneys have

dilated pelvicaliceal systems, either because dilatation has not yet developed, the upper

tract has decompressed by forniceal rupture (45), or the upper tract is noncompliant. In

addition, the diuretic effect of intravenous contrast likely explains why dilatation is

occasionally absent at US but present at IVU (45). Reports of the false negative rates

for US include absence of dilation in (1) 8 of 21 (38%) patients with a delayed

nephrogram on IVU (29); (2) 7 of 20 (35%) patients with clinically proven obstruction

(45); (3) 3 of 53 (6%) patients with IVU-proven obstruction (39); and (4) 5 of 14 (36%)

patients with IVU-proven obstruction (46).

The false negative rate of 6–38% for US in acute renal colic represents a major

limitation. Doppler US has been advocated as a method of identifying nondilated

obstructed kidneys, based on the increased vascular resistance in obstructed kidneys.

However, this is not widely used because the results have been disappointing and the

technique is operator dependent (46). The addition of forced diuresis has been sug-

gested (47), but is not in common use. Color flow Doppler may be useful in assessing

ureteral jets in the bladder. On average, 1–12 jets of urine arising from the ureteral

orifices will be observed per minute in an asymptomatic patient (Fig. 16). Ureteral jets

are usually symmetric in appearance. Asymmetry of visible ureteral jets on color flow

Doppler or asymmetric flow velocity on pulsed Doppler may suggest ureteral obstruc-

tion on the side of reduced flow. For this sign to be most useful, the patient must be

well hydrated. As some nonobstructed patients demonstrate variability in the pattern

and symmetry of ureteral jets, this technique is felt to be of limited usefulness.

Fig. 15. Ultrasound image in a patient with right-sided flank pain. The right ureter (U) is dilated to

the level of an echogenic stone (arrow) that is casting a characteristic acoustic shadow (asterisk).

Chapter 20 / Imaging of Urinary Stone Disease 387

Combined Plain Radiography and Ultrasound

Given the limitations of KUB and US individually, there is no good reason to believe

that the combination is superior to either modality alone. This is generally supported by

the literature, with the exception of one paper that claimed that the combination was

100% sensitive and specific (100%), because all stones missed by US were fortuitously

visible on KUB (48). A more plausible study reported a sensitivity of 80% and a speci-

ficity of 58% for a combination of both tests (29).

Computed Tomography

Although many stones are radiolucent on plain radiographs, essentially all stones are

sufficiently dense to be visible on CT. The only exceptions are the urinary stone/sludge

formation associated with indinavir (49) and the very rare pure matrix stone composed

of mucous and debris (50). Therefore, the terms radiopaque and radiolucent are only

appropriate for plain radiography, and should not be used in the context of CT. The

volumetric capability of spiral CT allows rapid acquisition of overlapping thin slices

without the possibility of missing small stones between slices. The advent of spiral CT

has revolutionized the imaging of urinary stone disease, and noncontrast spiral CT has

emerged as a highly accurate modality for the identification of ureteral calculi. In a

comparative study of IVU and noncontrast CT in 40 patients with flank pain, 28 with

stones proven by retrieval or passage, IVU had a sensitivity and specificity of 64% and

92%, compared to 100% and 92% for noncontrast CT, respectively (30). Another study

of stone detection by noncontrast CT reported a sensitivity of 97% and a specificity of

96%(51). False positives and negatives are still possible because stones may be mistaken

for phleboliths or vascular calcifications, and vice versa. Contrast can be administered

in difficult cases to resolve any doubts. Other advantages of CT include better stone

characterization and anatomic localization, improved depiction of nonstone pathology,

and speed. CT allows examination of the entire abdomen and pelvis in a single breath

hold in most patients on modern multi-slice helical scanners. NCCT acquisition times

Fig. 16. Axial Doppler ultrasound image of the bladder demonstrating a right ureteral jet (arrow).

On average, 1–12 jets of urine arising from the ureteral orifices will be observed per minute in

a normal person.

388 Breiman and Coakley

are less than a minute and usually less than 20 s on multislice CT scanners, approx

10 s with 16 slice helical scanners. Because contrast is not required, NCCT is particularly

useful in patients with contrast allergy or impaired renal function. As intravenous iodi-

nated contrast is not administered, NCCT does not require a physician or nurse in atten-

dance during scanning, increasing the accessibility of the examination after hours. NCCT

for urinary stones has been criticized as lacking a functional component, as compared to

IVU, because contrast is not administered. This may limit the ability to depict obstruc-

tion based on an asymmetric nephrogram or delayed excretion. However, the assessment

of the chronicity and severity of obstruction may not significantly contribute to manage-

ment decisions or to the success of therapy. With respect to the radiation dose delivered

by CT, the risks of low dose radiation are controversial; there is general agreement that

the lowest reasonable dose that allows a diagnostic study should be used (52).

OMPUTED-TOMOGRAPHY TECHNIQUE

In our institution, we scan patients on single, 4, 8, and 16 slice helical scanners. We

select an mA in a range of 100–450, commensurate with the size of the patient, 240 mA

on average, with a higher mA selected for larger patients. Automatic modulation of mA

is available on some CT scanners. This modulates the mA on a scan-by-scan basis as

needed to achieve a preset level of image quality, accounting for the thickness and

X-ray attenuating characteristics of the tissues present in the area included in the slice

volume. The mA (and radiation dose) would be lower through the upper abdomen, which

include the lung bases (which attenuate relatively little X-ray), than through the pelvis,

which includes the iliac bones (which greatly attenuate X-rays), for example. Reduction

of mA helps lower radiation dose, without compromising stone detection (52). When

using a 16 slice scanner, we obtain 3 mm contiguous scans with the patient in the prone

position, with a pitch of 0.9, a rotation time of 0.75 s per rotation and 120 kVp. No

intravenous or oral contrast is initially administered. Patients are scanned prospectively

in the prone position to aid in distinguishing a bladder stone from a stone impacted in the

interureteric ridge portion of the ureterovesical junction, which may simulate a bladder

stone on a supine image (53) (Fig. 17). In our institution, noncontrast scans obtained to

search for urinary calculi are performed prospectively in the prone position, avoiding the

need to rescan with the associated reduction in radiation dose, prolongation of the exami-

nation, or need to recall the patient if the need for prone images was not recognized while

the patient was still in the department.

With multi-slice helical CT scanners, it is also possible to retrospectively create

thinner slices from the original standard thicker source images. Retrospective targeted

high-resolution reconstructions decreasing pixel size and increasing spatial resolution

may also supplement routine scans when needed (Fig. 18). These retrospective recon-

structions can be performed without the physical presence of the patient, as long as

source image raw data is still available. On newer scanners, raw data is usually available

for 12–36 h, depending on the number of patients scanned and the complexity of their

examinations. As mentioned previously, multiplanar reformations (MPR) may also be

helpful in resolving diagnostic dilemmas. MPRs and three-dimensional reformations

require only image data and not raw source data, allowing manipulation years after the

original examination, presuming digital data has been archived.

Delayed post intravenous contrast excretory phase CT scans may, on occasion, be

necessary to further delineate small ureteral stones and to differentiate them from

phleboliths in close proximity to the ureter or to assess unexpected renal parenchymal

Chapter 20 / Imaging of Urinary Stone Disease 389

abnormalities suspected on initial noncontrast scans. As with an IVU, a dense nephrogram

associated with delayed excretion of contrast and dilatation of the collecting system is

often associated with ureteral obstruction on post-contrast scans.

OMPUTED-TOMOGRAPHY SIGNS OF UROLITHIASIS AND OBSTRUCTION

Ureteral obstruction is associated with secondary signs on CT. These include unilat-

eral hydroureter and/or hydronephrosis. Subtle degrees of collecting system distention

or perinephric fat stranding are most easily recognized in or about the upper or lower

poles of the kidney. Perinephric or periureteral fat stranding may also be encountered

Fig. 18. (A) Nonenhanced 5–mm-thick CT section obtained on a multidetector CT scanner

showing a questionable calcific density (arrow) in the left ureterovesical junction. (B) Thin

section small field-of-view image reconstructed from the same dataset confirms a small stone

(arrow) in the left ureterovesical junction. The ability to retrospectively create such targeted high-

resolution reconstructions (while the raw data remains available) is one of the advantages of

multidetector CT scanners.

Fig. 17. Nonenhanced CT image demonstrating the advantage of prone positioning in the evalu-

ation of urinary stones. The distal right ureter (black arrow) is dilated to the level of a small stone

(white arrow) impacted in the ureterovesical junction. With a scan obtained in the supine position,

it could be difficult to distinguish a stone in the ureterovesical junction protruding into the bladder

from a stone that has passed from the ureter and was lying freely within the bladder.