Download Nephrogenic Systemic Fibrosis and Its Impact on Abdominal Imaging

Note: This copy is for your personal non-commercial use only. To order presentation-ready
copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.
OVERVIEW
1565
Nephrogenic Systemic
Fibrosis and Its Impact
on Abdominal Imaging1
CME FEATURE
See accompanying
test at http://
www.rsna.org
/education
/rg_cme.html
LEARNING
OBJECTIVES
FOR TEST 2
After reading this
article and taking
the test, the reader
will be able to:
■■Describe
the features of nephrogenic
systemic fibrosis and
its association with
GBCA administration.
■■Identify
the risk
factors for nephrogenic systemic
fibrosis.
■■Discuss
ways to
minimize the chance
of nephrogenic systemic fibrosis while
still ensuring safe
diagnostic imaging.
TEACHING
POINTS
See last page
Martin R. Prince, MD, PhD • Hong Lei Zhang, MD • Joan C. Prowda,
MD • Marc E. Grossman, MD • David N. Silvers, MD
The objective of this article is to review the current knowledge about
nephrogenic systemic fibrosis (NSF) and how to prevent it. More than
300 cases of NSF in patients with severe chronic renal insufficiency or
acute renal failure or in patients undergoing dialysis have been reported
in the peer-reviewed literature, with an overwhelming majority occurring within weeks to months after injection of a gadolinium-based contrast agent (GBCA). Because administration of a high dose of a GBCA
is a primary risk factor and because most high-dose magnetic resonance
(MR) imaging applications involve abdominal imaging (eg, liver and
abdominal MR angiography), NSF cases have been associated with
abdominal MR imaging. Additional major risk factors for developing
NSF include proinflammatory conditions, failure to perform dialysis
promptly after GBCA administration, use of nonionic linear contrast
agents, hyperphosphatemia, and younger age. Recent recommendations
to use GBCAs with caution in patients with acute renal failure, patients
receiving dialysis, or patients with an estimated glomerular filtration rate
of less than 30 mL/min have resulted in virtually no new NSF cases being reported with onset in 2008 or 2009 in spite of a high level of awareness about this entity. In conclusion, NSF has been virtually eliminated
by using caution in administering GBCAs to patients known to have
severe or acute renal failure. In these patients, avoid high doses; and for
patients undergoing dialysis, schedule MR imaging to occur just before
a dialysis session to ensure rapid elimination of gadolinium.
©
RSNA, 2009 • radiographics.rsna.org
Abbreviations: FDA = Food and Drug Administration, GBCA = gadolinium-based contrast agent, GFR = glomerular filtration rate, NSF = nephrogenic systemic fibrosis, SNR = signal-to-noise ratio
RadioGraphics 2009; 29:1565–1574 • Published online 10.1148/rg.296095517 • Content Codes:
1
From the Departments of Radiology, Dermatology, and Dermatopathology, Weill Cornell Medical Center, Columbia College of Physicians and
Surgeons, 416 E 55th St, New York, NY 10022. Received April 14, 2009; revision requested May 20 and received June 15; accepted June 30.
M.R.P. has patent agreements with GE Healthcare, Hitachi, Siemens, Koninklijke Philips Electronics, Nemoto, Bayer, EPIX Pharmaceuticals,
Bracco Group, Covidien, and Topspins; all other authors have no financial relationships to disclose. Address correspondence to M.R.P. (e-mail:
[email protected]).
©
RSNA, 2009
1566 October Special Issue 2009
Introduction
Gadolinium-based contrast agents (GBCAs) have
proved to be among the safest available for clinical imaging. In particular, with the doses used for
magnetic resonance (MR) imaging, there are low
rates of occurrence of nephrotoxicity (1,2) and allergic reactions (3) compared with those rates for
iodinated contrast agents. The use of gadolinium
with MR imaging allows contrast enhancement
to be assessed without exposure to ionizing radiation. Until recently, these agents were used at
ever increasing doses in patients with renal failure
and patients undergoing dialysis. However, the
recent discovery of an association between GBCA
administration and nephrogenic systemic fibrosis
Teaching (NSF) has changed the way that gadolinium is
Point
used. High doses are now rare because most applications have been adjusted to use a standard dose
of 0.1 mmol/kg, especially in patients undergoing
dialysis or patients with a glomerular filtration rate
(GFR) of less than 30 mL/min. When patients are
undergoing dialysis, MR imaging with a GBCA is
scheduled for just before the next dialysis appointment to facilitate prompt clearance.
NSF is a rare fibrosing condition occurring in
patients with profound renal failure or patients undergoing dialysis. In NSF, patches of skin become
thickened and tethered to the underlying tissue,
reducing range of motion and leading to contractures. The fibrosing process can also involve internal organs, including the lungs, heart, and muscles
(4). Although many cases are mild and limited to
dermatologic manifestations, an estimated 5% of
cases have a more fulminant course resulting in
death (5). Because treatment options are limited,
an emphasis on prevention has been under way,
limiting GBCA exposure in patients with severe
renal failure (estimated GFR < 30 mL/min) and
dialysis patients. A virtual absence of new cases
with onset in 2008 or 2009 suggests that these
preventive measures have been successful and are
now allowing safe judicious use of GBCAs in patients with renal failure and dialysis patients when
clinically necessary (6).
The purpose of this review article is to outline
what is known about NSF and how to prevent it.
The article covers the history of gadolinium and
NSF, as well as the risk factors for NSF, its features, and its treatment. The impact of NSF on
abdominal imaging is discussed, and recommendations to prevent NSF are presented.
History of Gadolinium and NSF
Although gadolinium is a rare earth metal, with
the atomic number 64 on the periodic table, it is
actually common in the earth’s crust (7). When
radiographics.rsna.org
discovered and isolated in 1880, gadolinium was
named for the Finnish chemist Johan Gadolin, although it was actually discovered by Jean Charles
Galissard de Marignac and Paul Émile Lecoq de
Boisbaudran. Gadolinium has seven unpaired
electrons, making it one of the strongest paramagnetic atoms of the periodic table. Even when
bound to a chelator, unpaired electrons are available to interact with protons and other nearby
nuclei to facilitate longitudinal relaxation time
(T1) and enhance T1-weighted images. The susceptibility of gadolinium influences transverse relaxation time (T2), allowing perfusion and other
types of imaging that rely on gadolinium-induced
shortening of the transverse relaxation rate as affected by magnetic field inhomogeneity (T2*).
The first clinically successful chelator was
pentetic acid (diethylenetriaminepentaacetic
acid [DTPA]), an ionic linear agent (8) that was
borrowed from nuclear medicine and became
enormously popular as a gadolinium chelator
after receiving approval from the Food and Drug
Administration (FDA) in 1988. At that time,
nonionic iodinated agents were being discovered
to have safety advantages compared with ionic
iodinated agents (9). Accordingly, it was only a
matter of time before nonionic based contrast
agents were introduced in 1992 (gadoteridol) and
1993 (gadodiamide). As more and more reports
then emerged about the relative safety of GBCAs
in patients with renal failure compared with iodinated contrast agents, the use of GBCAs began
to increase in patients with renal failure and patients undergoing dialysis.
In the mid-1990s, greater awareness of the
relationship of the signal-to-noise ratio (SNR) to
the contrast agent dose led to increasing GBCA
use at high doses for SNR-limited applications
such as liver MR imaging and contrast-enhanced
MR angiography. The lower viscosity and lower
osmolarity, with lower irritation in the event of
extravasation, led to preferential use of nonionic
agents for the high-dose high-injection-rate applications. Body MR applications in the abdomen
commonly used GBCAs at high doses because
renal artery, portal vein, and liver MR imaging
were noted to benefit from the higher SNR of a
high GBCA dose (10).
The first known case of NSF occurred in 1997,
after the use of GBCAs at high doses in patients
with renal failure had become routine. At that
time, the entity was called nephrogenic fibrosing
dermopathy because of its manifestation primarily with skin symptoms. After the discovery of internal organ involvement, the name was changed
to reflect manifestations deep to the skin. The
cause of NSF remained mysterious until a breakthrough observation by Grobner (11) in Denmark in 2006. Grobner, a nephrologist, had five
RG ■ Volume 29 • Number 6
cases of NSF out of nine patients with renal failure whom he had sent for gadolinium-enhanced
MR imaging. Most of the radiologic community
was skeptical about this association until High
et al (12,13) and Wiginton et al (14) discovered
gadolinium in the biopsy specimens of NSF patients. An ensuing rush of articles reporting more
than 240 cases of NSF after GBCA administration has further buttressed this hypothesis.
After reviewing the scientific articles and in
consultation with the imaging and chemistry
experts, regulatory authorities in Europe and in
the United States issued a series of warnings culminating in a black-box warning label for all
GBCAs sold in the United States (15). This
warning recommends caution in the use of
GBCAs in patients with acute or chronic renal
dysfunction, a GFR of less than 30 mL/min, or
acute renal failure that is due to the hepatorenal
syndrome or in patients during the perioperative
period for liver transplantation. Risks and benefits of GBCAs should be assessed, and MR imaging without GBCAs or other alternate diagnostic
procedures should be considered in patients at
risk of NSF. Prompt hemodialysis after GBCA
administration should be considered for patients
already undergoing long-term therapy with hemodialysis. The European authorities have issued
similar warnings, as well as contraindications, for
certain GBCAs in patients undergoing dialysis or
patients with a GFR of less than 30 mL/min.
NSF Risk Factors
Renal Failure
Teaching
Point
NSF is only observed in patients with severe renal
dysfunction, primarily patients undergoing or approaching dialysis, hence the use of nephrogenic in
its name. Chronic renal failure has been categorized into multiple grades on the basis of the estimated GFR. GFR is typically estimated from the
Modification of Diet in Renal Disease (MDRD)
model that is based on age, gender, ethnicity, and
serum creatinine level. Online calculators are
readily available for determining GFR (16). It is
no longer considered appropriate to base clinical decision making about the risk of NSF on the
serum creatinine level alone. Most cases of NSF
are occurring in patients undergoing dialysis or
in patients with end-stage renal dysfunction, with
a GFR of less than 15 mL/min, who are not yet
undergoing dialysis. The published literature includes more than 300 cases of NSF, all in patients
with severe renal dysfunction and a GFR of less
than 30 mL/min (17). No known cases of NSF
have occurred in patients with a GFR of more
than 30 mL/min without acute renal failure. Accordingly, 30 mL/min has become the standard
cutoff value for determining whether a patient
Prince et al 1567
with chronic renal dysfunction could be at risk for
NSF. Furthermore, all patients who develop NSF
are known to have severe renal dysfunction. Accordingly, there is minimal, if any risk in patients
for whom the renal status is unknown.
It is possible to have acute renal failure with
a GFR of less than 30 mL/min but an erroneous
estimated GFR that is greater than 30 mL/min
because the serum creatinine level is increasing.
For this reason, the recommendation is to avoid
GBCA administration in patients with acute renal failure who have increasing serum creatinine
levels (18). Patients with acute renal failure can
undergo GBCA-enhanced MR imaging after
starting dialysis or after the creatinine level has
peaked and begun to decrease (18).
Originally, dialysis was believed to be a risk
factor for NSF because 70%–80% of the NSF patients were undergoing dialysis at the time of diagnosis. However, it now appears that some NSF patients may have received GBCAs prior to starting
dialysis. Therapy with dialysis may then have been
initiated by the time that NSF was diagnosed, reflecting the severity of the renal dysfunction at the
time of the GBCA administration. Furthermore,
patients who received GBCAs to evaluate a failing
dialysis fistula may not undergo adequate dialysis
after GBCA administration. In a more detailed
study, investigators have suggested that dialysis
performed the same day after GBCA administration is protective (18). Accordingly, for patients
undergoing dialysis who receive GBCAs, dialysis
should be performed as quickly as possible afterward. It is recommended to schedule MR imaging
to be performed just before a routine dialysis to
make sure that the dialysis is occurring well within
24 hours of any GBCA injection.
Teaching
Point
GBCA Dose
In multiple case series, investigators have reported that high doses of GBCAs, and possibly
Teaching
the use of linear nonionic GBCAs, contribute
Point
to an increased risk of NSF. Broome et al (19)
reported zero cases of NSF in 94 dialysis patients receiving the standard dose, 0.1 mmol/kg,
of gadodiamide, a linear nonionic agent; but 12
of 207 dialysis patients receiving a high dose of
gadodiamide developed NSF, with an odds ratio
of 12 to 1. In a larger but still retrospective study,
investigators found zero cases of NSF in approximately 74,000 unscreened patients receiving a
standard dose of a GBCA (80% received gadodiamide) but 15 cases of NSF in approximately
9000 patients receiving a high dose of a GBCA
(18). In case-control studies, Kallen et al (20),
Marckmann et al (21), and Collidge et al (22)
radiographics.rsna.org
1568 October Special Issue 2009
further implicated the use of high GBCA doses
as a major risk factor for NSF. In a comprehensive review of the literature, there were 180 cases
of NSF reported in which the gadolinium dose
could be estimated (23). In 157 (87%) of these
cases, a high dose was administered. These data
suggest that the risk of NSF can be reduced by
an order of magnitude simply by avoiding the use
of a high dose of more than 0.1 mmol/kg in patients undergoing dialysis or patients with a GFR
of less than 30 mL/min.
GBCA Type
Considerable debate surrounds the apparent differences in chelate stability between macrocyclic
and linear agents, as well as between ionic and
nonionic linear agents. Linear agents have an advantage compared with macrocyclic agents in that
they are simpler and generally less expensive to
synthesize. Macrocyclic agents, on the other hand,
have greater kinetic stability that is due to the
complete encircling of the Gd3+ ion with covalent
bonds. Indeed, the macrocyclic bond is so tight for
some macrocyclic agents that the kinetic dissociation time constant is on the order of months (24),
making it virtually impossible for Gd3+ ions to be
released into the tissue even when there is delayed
excretion. To the extent that the etiology of NSF
may be related to dissociation of the gadolinium
ion from the chelator, one can surmise that macrocyclic agents should be completely safe and essentially free of risk of NSF. Indeed, there are few
cases of NSF associated with macrocyclic agents,
although because of the substantially lower market
share of these contrast agents, caution should be
exercised about assuming that macrocyclic agents
are completely without risk.
Another controversy surrounds the differences between ionic and nonionic agents. It is well
known with iodinated contrast agents that nonionic agents are safer, with a far lower incidence
of allergic reactions (25). However, it appears that
the in vitro stability of the nonionic linear agents
is less than that for ionic agents, and there are
more cases of NSF with nonionic agents, even
after correcting for market share. One wonders if
in vivo stability of the nonionic agents may also be
lower and that this may contribute to the increased
risk of NSF. Because of these theoretical reasons,
the use of linear nonionic agents, in spite of their
favorable adverse-event profile with a low risk of
life-threatening allergic reactions (3), has not been
recommended, especially at high doses, in patients
with a GFR of less than 30 mL/min (26).
Proinflammatory Events
In their case series, Sadowski et al (27) noticed
that all NSF cases were occurring in relatively
sick inpatients. No cases of NSF were found in
outpatients. All of the NSF patients had underlying proinflammatory conditions, which were
defined as recent major surgery, infection, or
vascular event. In the case-control analysis, these
investigators demonstrated a significantly greater
number of proinflammatory conditions per patient in NSF patients (P < .001) compared with
similar control patients who did not develop
NSF. This observation is consistent with a prevalent theory about the etiology of NSF. In this theory, fibrocytes circulating in the blood because of
an ongoing proinflammatory event are attracted
to where gadolinium has deposited in the skin
and other organs. This attraction may be due to
phagocytosis of gadolinium by macrophages and
to the release of cytokines that attract circulating
fibrocytes. If there is no ongoing intense inflammation in the body, there will be insufficient fibrocytes in circulation to cause NSF.
In addition, proinflammatory conditions are
associated with edema, which locally increases
the extravascular-extracellular compartment. In
theory, the expansion of the interstitial compartment may give edematous areas slower egress of
the GBCA, with more time to cause NSF.
Age
In numerous articles, investigators have reported
that younger age is associated with a higher incidence of NSF in the at-risk population with a
GFR of less than 30 mL/min (18,27). This may
be partly explained by inaccuracies in the Modification of Diet in Renal Disease (MDRD) model’s
estimation of GFR in older patients. However,
the peak incidence of NSF occurs in patients
around 50–60 years of age, even though patients
older than 60 have a substantially higher rate of
GBCA-enhanced MR imaging studies and renal
disease. One hypothesis is that this observation
reflects the stronger more-active immune system
of younger patients and the reduced collagen synthesis in older patients (23).
Hyperphosphatemia
One theory about the cause of NSF relates to
gadolinium ions dissociating from the chelator.
Normally, Gd3+ would quickly reassociate with
the chelator. However, if Gd3+ irreversibly binds
to phosphates and precipitates out of solution
into tissues, then the reaction releasing gadolinium from the chelator becomes unidirectional.
This unidirectional reaction will accelerate in
RG ■ Volume 29 • Number 6
Prince et al 1569
Figure 1. Photographs of skin lesions in NSF, including skin thickening with erythematous plaques (a)
and thickened skin with hyperpigmented plaques (b).
patients who have more phosphate in the serum.
Indeed, a higher incidence of NSF has been associated with hyperphosphatemia (23).
Epoetin, Acidosis, and Other Factors
In various case series, investigators have reported
a number of additional associations that have not
been consistently corroborated. Epoetin is commonly used to treat anemia in patients undergoing
dialysis and patients with renal failure. Because
epoetin stimulates the bone marrow relatively
indiscriminately, including both red and white
blood cells, it is considered to be proinflammatory. In several cases series, investigators have
noted a high percentage of NSF patients who are
receiving epoetin, including “high-dose” epoetin
therapy (28,29). Dissociation of gadolinium from
the chelator is known to occur more readily in an
acid environment, where there are more positively
charged protons competing for the gadolinium
ionic binding sites. In several case series, including the first report by Grobner (11), investigators
have noticed a high prevalence of acidosis in the
patients with NSF (14,30). However, acidosis is
common in patients with renal failure. Additional
factors hypothesized to have a relationship to NSF
include elevated concentrations of serum cations,
including iron, zinc, and calcium.
Comparison to Other Radiologic Risks
Although the historical risk of NSF with high-dose
GBCA administration (mainly linear nonionic
agents) has been reported to be in the 1%–7%
range for patients undergoing dialysis, this risk
has now been reduced by orders of magnitude by
the current practice of using the standard dose,
0.1 mmol/kg, and having patients undergo dialysis
immediately after GBCA administration. The use
of more-stable agents may also contribute to less
risk. At the hospitals of our institution, more than
100,000 single-dose GBCA administrations have
now been given, including more than 10,000 in
patients with a GFR of less than 30 mL/min, with
no cases of NSF developing. This risk is less than
the risk of nephrotoxicity, which has been reported
to be as high as 50%–90% in patients with renal
insufficiency (31–33).
NSF Features
Clinical Manifestation
NSF is different from the contrast reactions usually encountered by radiologists because it does
not occur at the time of the imaging study. Instead, NSF typically occurs weeks to months later.
NSF lesions usually involve the dermis in the
extremities symmetrically and, less commonly, the
trunk. The face is commonly spared, an important
differential diagnostic point that allows clinical
distinction from other entities with hard skin (34).
The tempo of disease onset is variable. Symptoms
and signs of NSF may be overlooked by both the
patient and the physician or may rarely be rapidly
progressive (fulminant) with systemic involvement.
It is more common for patients to have a moreindolent disease course, with several weeks to
months of symptoms before diagnosis.
The clinical manifestation of NSF more closely
resembles scleromyxedema or eosinophilic fasciitis. The primary lesions are firm to hard skincolored or erythematous papules and plaques
that may coalesce to involve large areas of the
trunk or extremity. The skin lesions are described
as a hyperpigmented thickened brawny induration that is woody or edematous or has a peau
d’orange or cobblestoned appearance (Fig 1) (34).
radiographics.rsna.org
1570 October Special Issue 2009
Figures 2–4. (2) Axial CT image shows trunk and
breast skin thickening in a patient with biopsy-confirmed NSF. (3) Axial CT image of the left thigh shows
skin thickening and stranding in the subcutaneous fat
in a region of dermal NSF. (4) Coronal MR image
of thighs of a patient with NSF shows subcutaneous
edema, skin thickening, and deep dermal stranding
from thickened fibrous bands.
These plaques typically spare the antecubital and
popliteal fossae and have advancing ameboid projections on the distinctive irregular edge. NSF is
characterized by progressive skin hardening and
tethering. Joint contractures may develop when
the fibrosis spans the joint and may severely impair
physical function, including ambulation. The flexion contractures may progress to the point of patients becoming wheelchair bound. Patients commonly complain of severe extremity pain, pruritus,
skin tightness, or a burning sensation.
Radiologic Manifestation
Because of the nonspecific symptoms of NSF,
these patients commonly undergo radiologic
studies, which may show (a) diffuse soft-tissue
tracer uptake on bone scans, (b) skin thickening
and subcutaneous stranding on mammographic,
computed tomographic (CT) (Figs 2, 3), MR
(Fig 4), and ultrasonographic (US) images, and
(c) skin and muscle activity on positron emission tomographic images (35). NSF is not vis-
ible at digital subtraction angiography, and the
practice of using high-dose GBCAs in place of
iodine-based contrast agents for digital subtraction angiography in patients with renal failure can
no longer be justified until more is learned about
preventing and treating NSF.
Histologic Findings
Because of the involvement of the subcutaneous tissues and underlying muscle in NSF, a
deep dermal biopsy is necessary to confirm the
diagnosis histologically. However, because the
histopathologic finding is nonspecific inflamma-
RG ■ Volume 29 • Number 6
Prince et al 1571
Impact on Abdominal Imaging
Figure 5. Three-dimensional nonenhanced MR
angiographic image of a renal transplant recipient obtained at 1.5 T (Enhance pulse sequence; GE Healthcare, Chalfont Saint Giles, United Kingdom).
tory changes, biopsy can be used only to confirm
a clinical diagnosis. Both characteristic clinical
features and the biopsy findings together are
necessary for definitive diagnosis (4). Histologic
specimens show thickened collagen bundles, mucin deposition, spindle cell proliferation, and fibroblasts that stain for CD3, indicating that they
are circulating fibrocytes.
Treatment
There is no consistently effective therapy for NSF.
Improving renal dysfunction from any cause appears to slow or arrest the progression of NSF. In
numerous case series, investigators have reported
instances of cure or a reduction of symptoms after
successful renal transplantation or recovery from
acute renal failure (11,36,37). Physical therapy
should be started early to maintain and improve
the range of motion of contracted joints. Multiple
case reports of different therapeutic interventions
have appeared since the first published report of
NSF by Cowper et al (38). Extracorporeal photopheresis (39), plasmapheresis, photodynamic
therapy, pentoxifylline therapy, high-dose intravenous immunoglobulin therapy, thalidomide
therapy, steroid therapy, and other immunosuppressive therapies have been attempted. The latest
additions to the list of treatments include imatinib
mesylate (Gleevec; Novartis, East Hanover, NJ)
(40) and sodium thiosulfate.
One of the areas of greatest impact of NSF has
been the imaging of patients with renal failure,
especially a patient with a kidney transplant
who is undergoing MR angiography to look for
renal artery stenosis (Fig 5). Previously, these
patients had been imaged with a high dose of a
GBCA. However, now a number of techniques
that do not use contrast agents may be able
to be used to assess the renal arteries without
GBCA administration, including two- or threedimensional steady-state free precession (41),
three-dimensional phase contrast (42), and a
more advanced version of time of flight that allows a longer time for inflow after an inversion
pulse (43). When these fail and three-dimensional
GBCA-enhanced MR angiography is still necessary in a patient with a GFR of less than 30 mL/
min, it can be performed on state-of-the-art MR
imagers by using no more than a standard dose
of 0.1 mmol/kg. Some benefit may be seen from
higher-relaxivity agents (44), including gadofosveset trisodium, which has recently received FDA
approval for use in MR angiography at a dose of
only 0.03 mmol/kg (45).
Hepatic imaging has also been impacted because high-dose GBCA-enhanced MR imaging
had been increasingly used to characterize hepatic lesions and detect hepatocellular carcinoma
in patients with cirrhosis. It is now possible to get
diagnostic liver MR imaging with standard-dose
GBCA administration by using a state-of-the-art
MR imager with high-SNR phased-array coils
and highly optimized three-dimensional fatsuppressed spoiled gradient-echo sequences (eg,
liver acquisition with volume acceleration [LAVA;
GE Healthcare], volumetric interpolated breathhold examination [VIBE; Siemens, Erlangen,
Germany], T1-weighted high-resolution isotropic
volume examination [THRIVE; Philips, Andover,
Mass]). Because using the standard dose of 0.1
mmol/kg reduces risk by more than an order of
magnitude, most protocols have adopted this
standard dose for liver MR imaging and portal
venous MR angiography. Similarly, for characterizing renal masses by imaging before and after
GBCA injection, a standard dose of 0.1 mmol/kg
helps to minimize risk.
1572 October Special Issue 2009
Some have advocated using less than the standard dose. This is possible with gadoxetic acid
(gadoxetate disodium [Eovist; Bayer Healthcare,
Wayne, NJ]), which has now received FDA approval for liver imaging at a dose of 0.025 mmol/
kg. Gadoxetic acid comes with a reduced concentration, half that of regular GBCAs, so the
injection rate needs to be high, preferably 3 mL/
sec. However, the bolus will then be short, so it is
helpful to use a short arterial phase of 15–20 seconds or less and to use fluoro-triggering or some
other bolus timing and/or triggering method to
ensure precise synchronization of the arterial
phase with acquisition of the center of k-space.
In patients with profound renal insufficiency
with a GFR of less than 15 mL/min or patients
with acute renal failure who are not undergoing dialysis, diffusion-weighted MR imaging
has been proposed as an alternative to contrastenhanced sequences. On the apparent diffusion
coefficient images, malignant liver tumors have
higher diffusivity than benign lesions do (46,47).
Although diffusion-weighted MR imaging is difficult in the abdomen, where breathing motion
confuses images of molecular motion, the use
of echo-planar techniques has substantially improved image quality.
In some instances, it may be possible to switch
to another type of examination in patients at risk
of NSF. If US or nonenhanced CT can resolve the
clinical issue, then these modalities can be used.
However, because the risk of NSF is less than
the risk of nephrotoxicity from iodinated contrast
agents and because the risk of death from NSF is
less than the risk of death from allergic reactions
to iodinated contrast agents, it does not make
sense to switch patients to contrast-enhanced CT
just to avoid NSF. In addition, the use of iodinated
contrast agents in patients undergoing dialysis
should not be considered lightly because this may
destroy any remaining nephrons that are able to
sustain the patient between dialysis treatments.
Recommendations to Prevent NSF
Because there is no consistently effective treatment for NSF, prevention is important. Screening
to identify patients at high risk is essential (48).
Considerable controversy remains with regard
radiographics.rsna.org
to what level of screening is necessary. Because
70%–80% of NSF cases occur in patients undergoing dialysis, screening for dialysis is important.
Dialysis patients should be scheduled to undergo
their MR imaging just before dialysis so that if
administration of a GBCA is necessary, it will be
quickly removed after MR imaging. Data on the
serum creatinine level should be looked up for all
inpatients, and the GFR should be calculated to
identify patients with a GFR of less than 30 mL/
min. The GFR calculation can be done with any
of a number of online calculators (16).
Outpatients who are not undergoing dialysis
rarely get NSF and generally do not have recent
data on the serum creatinine level available, so
the screening procedures for outpatients are
more controversial. Because 20% of NSF cases
occur in patients with renal transplants, these
outpatients should have the serum creatinine
level checked to calculate the GFR. Usually it
is not necessary to draw blood because kidney
transplantation patients are having their serum
creatinine level checked on a regular basis. All
outpatients should be asked if they have renal
disease, and any outpatients responding positively
should have the serum creatinine level checked.
Because patients with profound liver disease in
the peritransplantation period may readily develop hepatorenal failure, it is important to check
the serum creatinine level in these patients as
well. Because it is necessary to have a proinflammatory condition together with severe renal failure in order to develop NSF and because patients
with proinflammatory conditions generally have
undergone recent serum creatinine testing, drawing blood to check the serum creatinine level is
rarely necessary.
For any patients with a calculated GFR of
less than 30 mL/min, the precontrast MR images should be checked before giving the contrast agent to be sure GBCA use is necessary. If
GBCA administration is necessary, a maximum
of the standard dose, 0.1 mmol/kg, should be
used. Hydration has been recommended to
help with GBCA elimination after administration. Some authors have suggested obtaining
informed consent prior to administering the
GBCA. Informed consent is complicated by the
need to provide patients with an accurate estimate of the risk of NSF. At our hospital, with no
screening for renal function, the risk of NSF is
RG ■ Volume 29 • Number 6
less than one in 100,000 and is estimated to be
less than one in 10,000 for patients with a GFR
of less than 30 mL/min with the 0.1-mmol/kg
dose. This risk is similar to the risk of allergic
reaction to GBCA (3), for which we do not routinely obtain informed consent.
There is only one contrast agent, gadofosveset trisodium, that has FDA approval for
use in MR angiography, and this agent is only
just becoming commercially available. To our
knowledge, no known cases of NSF have been
reported with this contrast agent, and it has six
times the relaxivity of standard GBCAs. High
relaxivity offers the possibility of high-SNR MR
angiography with a low dose. In the future, it
may be possible to perform contrast-enhanced
MR angiography with iron oxide–based contrast
agents, which would completely eliminate the
risk of gadolinium-induced NSF (18).
Conclusion
Teaching
Point
Although the discovery of NSF has shattered the
safety reputation of GBCAs, understanding the
NSF risk mechanisms now allows safe use of lowdose GBCA administration in most patients.
References
1.Prince MR, Arnoldus C, Frisoli JK. Nephrotoxicity
of high-dose gadolinium compared with iodinated
contrast. J Magn Reson Imaging 1996;6:162–166.
2.Rofsky NM, Weinreb JC, Bosniak MA, Libes RB,
Birnbaum BA. Renal lesion characterization with
gadolinium-enhanced MR imaging: efficacy and
safety in patients with renal insufficiency. Radiology
1991;180:85–89.
3.Murphy KJ, Brunberg JA, Cohan RH. Adverse reactions to gadolinium contrast media: a review of 36
cases. AJR Am J Roentgenol 1996;167:847–849.
4.Cowper SE. Nephrogenic systemic fibrosis: an overview. J Am Coll Radiol 2008;5:23–28.
5.Cowper SE. The International Center for Nephrogenic Fibrosing Dermopathy Research (ICNFDR).
ICNFDR Web site. http://www.icnfdr.org. Updated
June 18, 2009. Accessed August 25, 2009.
6.Leiner T. Current insights into NSF [abstr]. In: Proceedings of the 16th Meeting of the International
Society for Magnetic Resonance in Medicine. International Society for Magnetic Resonance in Medicine Web site. http://cds.ismrm.org/ismrm-2008
/isv7/main.htm. Accessed August 25, 2009.
7.Gadolinium. Wikipedia Web site. http://en.wikipedia
.org/wiki/Gadolinium. Updated July 31, 2009. Accessed August 25, 2009.
Prince et al 1573
8.Weinmann HJ, Brasch RC, Press WR, Wesbey GE.
Characteristics of gadolinium-DTPA complex: a potential NMR contrast agent. AJR Am J Roentgenol
1984;142:619–624.
9.Yamaguchi K, Katayama H, Takashima T, Kozuka
T, Seez P, Matsuura K. Prediction of severe adverse
reactions to ionic and nonionic contrast media in
Japan: evaluation of pretesting—a report from the
Japanese Committee on the Safety of Contrast Media. Radiology 1991;178:363–367.
10.Schneider G, Prince MR, Meaney JFM, Ho VB.
Magnetic resonance angiography: techniques, indication and practical applications. Milan, Italy:
Springer-Verlag Italia, 2005.
11.Grobner T. Gadolinium—a specific trigger for the
development of nephrogenic fibrosing dermopathy
and nephrogenic systemic fibrosis? Nephrol Dial
Transplant 2006;21:1104–1108.
12.High WA, Ayers RA, Chandler J, Zito G, Cowper
SE. Gadolinium is detectable within the tissue of
patients with nephrogenic systemic fibrosis. J Am
Acad Dermatol 2007;56:21–26.
13.High WA, Ayers RA, Cowper SE. Gadolinium is
quantifiable within the tissue of patients with nephrogenic systemic fibrosis. J Am Acad Dermatol
2007;56:710–712.
14.Wiginton CD, Kelly B, Oto A, et al. Gadoliniumbased contrast exposure, nephrogenic systemic fibrosis, and gadolinium detection in tissue. AJR Am J
Roentgenol 2008;190:1060–1068.
15.Information for healthcare professionals gadolinium-based agents for magnetic resonance imaging
(marketed as Magnevist, MultiHance, Omniscan,
OptiMARK, ProHance). Food and Drug Administration Web site. http://www.fda.gov/Drugs/Drug
Safety/PostmarketDrugSafetyInformationforPatient
sandProviders/ucm/142884.htm. Published June
2006. Updated August 13, 2009. Accessed August
25, 2009.
16.Fadem SZ. MDRD GFR calculator (with SI units):
4 variable MDRD study equation using serum creatinine, age, race, gender. Nephron Information Center Web site. http://www.mdrd.com. Updated August
19, 2009. Accessed August 25, 2009.
17.Willicombe M, Cunningham J. Nephrogenic systemic fibrosis: a sufficient reason to avoid gadolinium-based contrast in all patients with renal impairment? Semin Dial 2008;21:140–141.
18.Prince MR, Zhang H, Morris M, et al. Incidence of
nephrogenic systemic fibrosis at two large medical
centers. Radiology 2008;248:807–816.
19.Broome DR, Girguis MS, Baron PW, Cottrell AC,
Kjellin I, Kirk GA. Gadodiamide-associated nephrogenic systemic fibrosis: why radiologists should
be concerned. AJR Am J Roentgenol 2007;188:
586–592.
1574 October Special Issue 2009
20.Kallen AJ, Jhung MA, Cheng S, et al. Gadoliniumcontaining magnetic resonance imaging contrast
and nephrogenic systemic fibrosis: a case-control
study. Am J Kidney Dis 2008;51:966–975.
21.Marckmann P, Skov L, Rossen K, Heaf JG, Thomsen HS. Case-control study of gadodiamide-related
nephrogenic systemic fibrosis. Nephrol Dial Transplant 2007;22:3174–3178.
22.Collidge TA, Thomson PC, Mark PB, et al. Gadolinium-enhanced MR imaging and nephrogenic systemic fibrosis: retrospective study of a renal replacement therapy cohort. Radiology 2007;245:168–175.
23.Prince MR, Zhang HL, Roditi GH, Leiner T,
Kucharczyk W. Risk factors for NSF: a literature
review. J Magn Reson Imaging (in press).
24.Schmitt-Willich H. Stability of linear and macrocyclic gadolinium based contrast agents. Br J Radiol
2007;80:581–582.
25.Katayama H, Yamaguchi K, Kozuka T, Takashima T,
Seez P, Matsuura K. Adverse reactions to ionic and
nonionic contrast media: a report from the Japanese
Committee on the Safety of Contrast Media. Radiology 1990;175:621–628.
26.American College of Radiology. ACR manual on
contrast media, version 6: 2008. American College
of Radiology Web site. http://www.acr.org
/SecondaryMainMenuCategories/quality_safety
/RadSafety.aspx. Published May 2008. Updated
October 6, 2008. Accessed August 25, 2009.
27.Sadowski EA, Bennett LK, Chan MR, et al. Nephrogenic systemic fibrosis: risk factors and incidence
estimation. Radiology 2007;243:148–157.
28.Peak AS, Sheller A. Risk factors for developing gadolinium-induced nephrogenic systemic fibrosis. Ann
Pharmacother 2007;41:1481–1485.
29.Grobner T, Prischl FC. Patient characteristics and
risk factors for nephrogenic systemic fibrosis following gadolinium exposure. Semin Dial 2008;21:
135–139.
30.Marckmann P. An epidemic outbreak of nephrogenic systemic fibrosis in a Danish hospital. Eur J
Radiol 2008;66:187–190.
31.Gleeson TG, Bulugahapitiya S. Contrast-induced
nephropathy. AJR Am J Roentgenol 2004;183:
1673–1689.
32.Thomsen HS. European Society of Urogenital
Radiology guidelines on contrast media application. Curr Opin Urol 2007;17:70–76.
33.Parfrey PS, Griffiths SM, Barrett BJ. Contrast material-induced renal failure in patients with diabetes
mellitus, renal insufficiency, or both: a prospective
controlled study. N Engl J Med 1989;320:143–149.
34.Cowper SE, Boyer PJ. Nephrogenic systemic fibrosis:
an update. Curr Rheumatol Rep 2006;8:151–157.
radiographics.rsna.org
35.Morris MF, Zhang Y, Zhang HL, et al. Features of
nephrogenic systemic fibrosis on radiology examinations. AJR Am J Roentgenol 2009;193:61–69.
36.Weigle JP, Broome DR. Nephrogenic systemic fibrosis: chronic imaging findings and review of the medical literature. Skeletal Radiol 2008;37:457–464.
37.Marckmann P, Skov L, Rossen K, et al. Nephrogenic systemic fibrosis: suspected causative role of
gadodiamide used for contrast-enhanced magnetic
resonance imaging. J Am Soc Nephrol 2006;17:
2359–2362.
38.Cowper SE, Su LD, Bhawan J, Robin HS, LeBoit
PE. Nephrogenic fibrosing dermopathy. Am J Dermatopathol 2001;23:383–393.
39.Mathur K, Morris S, Deighan C, Green R, Douglas
KW. Extracorporeal photopheresis improves nephrogenic fibrosing dermopathy/nephrogenic systemic
fibrosis: three case reports and review of literature. J
Clin Apher 2008;23:144–150.
40.Kay J, High WA. Imatinib mesylate treatment of
nephrogenic systemic fibrosis. Arthritis Rheum
2008;58:2543–2548.
41.Maki JH, Wilson GJ, Eubank WB, Glickerman DJ,
Pipavath S, Hoogeveen RM. Steady-state free precession MRA of the renal arteries: breath-hold and
navigator-gated techniques vs. CE-MRA. J Magn
Reson Imaging 2007;26:966–973.
42.Gu T, Korosec FR, Block WF, et al. PC VIPR: a
high-speed 3D phase-contrast method for flow
quantification and high-resolution angiography.
AJNR Am J Neuroradiol 2005;26:743–749.
43.Utsunomiya D, Miyazaki M, Nomitsu Y, et al. Clinical role of non-contrast magnetic resonance angiography for evaluation of renal artery stenosis. Circ J
2008;72:1627–1630.
44.Juluru K, Vogel-Claussen J, Macura KJ, Kamel IR,
Steever A, Bluemke DA. MR imaging in patients at
risk for developing nephrogenic systemic fibrosis: protocols, practices, and imaging techniques to maximize
patient safety. RadioGraphics 2009;29:9–22.
45.FDA approves first imaging agent to enhance scans
of blood flow. Food and Drug Administration Web
site. http://www.fda.gov/NewsEvents/Newsroom
/PressAnnouncements/2008/ucm117000.htm. Published December 24, 2008. Updated June 18, 2009.
Accessed August 25, 2009.
46.Zhang J, Tehrani YM, Wang L, Ishill NM, Schwartz
LH, Hricak H. Renal masses: characterization with
diffusion-weighted MR imaging—a preliminary experience. Radiology 2008;247:458–464.
47.Kim S, Naik M, Sigmund E, Taouli B. Diffusionweighted MR imaging of the kidneys and the urinary tract. Magn Reson Imaging Clin N Am 2008;
16:585–596.
48.Kribben A, Witzke O, Hillen U, Barkhausen J, Daul
AE, Erbel R. Nephrogenic systemic fibrosis: pathogenesis, diagnosis, and therapy. J Am Coll Cardiol
2009;53:1621–1628.
This article meets the criteria for 1.0 credit hour in category 1 of the AMA Physician’s Recognition Award. To obtain
credit, see accompanying test at http://www.rsna.org/education/rg_cme.html.
RG
Volume 29 • October Special Issue 2009
Prince et al
Nephrogenic Systemic Fibrosis and Its Impact on Abdominal
Imaging
Martin R. Prince, MD, PhD, et al
RadioGraphics 2009; 29:1565–1574 • Published online 10.1148/rg.296095517 • Content Codes:
Page 1566
However, the recent discovery of an association between GBCA administration and nephrogenic
systemic fibrosis (NSF) has changed the way that gadolinium is used. High doses are now rare
because most applications have been adjusted to use a standard dose of 0.1 mmol/kg, especially in
patients undergoing dialysis or patients with a glomerular filtration rate (GFR) of less than 30
mL/min.
Page 1567
NSF is only observed in patients with severe renal dysfunction, primarily patients undergoing or
approaching dialysis, hence the use of nephrogenic in its name.
Page 1567
For patients undergoing dialysis who receive GBCAs, dialysis should be performed as quickly as
possible afterward. It is recommended to schedule MR imaging to be performed just before a routine
dialysis to make sure that the dialysis is occurring well within 24 hours of any GBCA injection.
Page 1567
High doses of GBCA, and possibly the use of linear nonionic GBCAs, contribute to an increased risk
of NSF.
Page 1573
Although the discovery of NSF has shattered the safety reputation of GBCAs, understanding the NSF
risk mechanisms now allows safe use of low-dose GBCA administration in most patients.