is a third-year Resident in Diagnostic Radiology at Mallinckrodt
Institute of Radiology at the Washington University School of
Medicine, St. Louis, MO. He received a BS in Chemical Engineering
from the University of Virginia, Charlottesville, VA. He worked
as a Product-Development Engineer at Procter and Gamble for three
years. He received his MD from Ohio State University, Columbus,
OH in 1999. He will begin a Fellowship in Interventional
Radiology at Mallinckrodt in July 2004.
Gadolinium contrast media is typically used for its
paramagnetic properties in magnetic resonance imaging (MRI).
However, it can also be used in a completely different context
for its X-ray attenuation properties. Gadolinium contrast may be
substituted for iodinated contrast media in a wide range of
arterial and venous interventional procedures when combined with
digital subtraction angiography. This may be useful for patients
with a contraindication to the use of iodinated contrast, but it
should be stressed that this represents an off-label use of
gadolinium. A relatively small volume of contrast media may be
used if conventional MRI dosing is used as a guideline. Another
disadvantage is that vascular opacification is somewhat inferior
to that achieved with iodinated contrast.
It is not uncommon for patients requiring vascular
interventional procedures to have a contraindication to the use of
iodinated contrast. The two most common contraindications are a
history of an allergic reaction to iodinated contrast and renal
insufficiency. In such cases, gadolinium contrast media and
conventional digital subtraction angiography (DSA) may be a safe
and effective alternative for a broad range of procedures,
including evaluation of the renal and carotid arteries (Figure 1).
Gadolinium contrast media is a low-viscosity liquid that is used
in the same manner as iodinated contrast media with either hand or
power injection. In general, it provides good image quality, and in
many situations it is able to provide diagnostic information
equivalent to that obtained with the use of iodine. Image quality
is superior to that obtained using carbon dioxide (CO
) contrast, and it may also be used for procedures above the
diaphragm where the use of CO
is controversial or contraindicated.
Drawbacks of iodinated contrast
Among patients with a creatinine level >2.0 mg/dL, there is a
20% incidence of worsening renal function
following the use of iodinated contrast.
Risk factors for contrast-induced nephropathy include pre-existing
renal insufficiency, diabetes, multiple myelo- ma, congestive heart
failure, dehydration, and a high dose of contrast. Although most
cases of contrast-induced nephropathy resolve, the problem is not
trivial. Prolonged hospital stays are common, and 10% to 25% of
these patients may require dialysis.
Even though the incidence of severe adverse reactions to
iodinated contrast is only 10 per million,
it is a dreaded occurrence. Patients with a convincing history of
anaphylaxis should not be given this contrast. Patients with a
history of a milder reaction, such as a rash, should be pretreated
with corticosteroids prior to any future iodinated contrast
administrations. In the clinical setting, it is common for a
patient to report a history of allergy without recalling specific
details. Therefore, even though the type of reaction may not be
reliably known, the potential threat of a severe reaction must be
taken seriously. Other less common reasons for avoiding iodinated
contrast usage include hyperthyroidism, upcoming radioiodine
therapy or thyroid scintigraphy, metformin usage that cannot easily
be held, and concern for nausea in a patient who has not been
Compton scattering provides the majority of image contrast for
soft tissues in most diagnostic imaging studies. The photoelectric
effect becomes significant for X-ray beam attenuation from contrast
materials. The probability of this interaction increases with
atomic number of the target, which is higher for gadolinium (Z =
64) than for iodine (Z = 53). A marked increase in photon
absorption occurs at energies just above the "K-edge" of the target
material. The K-edge of gadolinium is 50 keV, and that of iodine is
33 keV. In practical terms, this means that a higher energy setting
should be used when performing imaging studies with gadolinium
contrast. The average energy of an X-ray beam is about one half of
the kVp, so to maximize image contrast, a setting of 90 to 100 kVp
is recommended for gadolinium studies.
This higher than normal energy setting also reduces the patient
radiation dose by 20% to 40%. In vivo, beam hardening from soft
tissues preferentially filters out photons with energies near the
K-edge of iodine, and, depending on beam energy and filtration,
gadolinium attenuates up to 50% more photons than iodine when
considered on an equimolar basis.
An important disadvantage of gadolinium in this setting,
however, is the low concentration of available preparations.
Iodinated contrast provides excellent vascular image quality at its
commonly used concentration of approximately 300 mg I/mL.
Concentration of iodinated contrast media is generally expressed in
milligrams of iodine per milliliter, while gadolinium contrast is
described in millimoles of gadolinium per milliliter. A solution of
300 mg I/mL is equal to 2.36 mmol I/mL. By comparison, gadolinium
contrast media is available at a maximum concentration of only 0.50
mmol/mL. After accounting for optimized beam energy, beam
hardening, and other practical considerations, it is estimated that
nondiluted gadolinium at 0.5 mmol/mL achieves the same image
contrast as 60 to 80 mg I/mL,
or roughly one-fifth of a standard 300 mg I/mL solution. Digital
subtraction angiography is needed to provide adequate vascular
contrast effect at this concentration. Image contrast using
conventional fluoroscopy is unsatisfactory, often even for test
runs. Additionally, dilution of gadolinium media with saline is not
recommended since this would cause further reduction in image
A second important disadvantage of gadolinium is the relatively
low maximum approved dose--0.3 mmol/kg of body weight--although,
strictly speaking, there is no approved dose for usage with DSA
since this is an off-label application. At a concentration of 0.5
mmol/mL, 0.3 mmol/kg provides 45 mL of media per examination for a
typical 75-kg patient. While this may be sufficient for a
relatively limited exam such as an inferior vena cavogram, it may
not be sufficient for a complete pelvic and lower extremity
arterial study. Modified techniques may be used in situations where
the maximum volume of gadolinium contrast is insufficient for a
complete study. For example, CO
imaging can be performed initially with gadolinium reserved for
clarifying specific lesions.
As mentioned previously, it is important to recognize that use
of gadolinium for DSA constitutes an off-label application.
Gadolinium is FDA-approved only for intravenous (IV) administration
for MRI, and, furthermore, is approved only for particular body
parts--no agent is expressly approved for cardiac imaging.
This fact alone should not dissuade the radiologist from
considering proven, but not approved, uses of gadolinium. Once a
product is approved for a particular use, a physician may see fit
to use that product in other ways that he believes will benefit the
For example, the intra-articular injection of gadolinium for
evaluation of joints is not FDA-approved but is a commonplace
procedure. However, the patient should be informed of the nature of
any off-label usages, and this discussion should be incorporated
into the informed consent process.
Of course, the use of IV gadolinium agents is standard practice
in MRI and, in that context, these agents have been found to be
very safe, with little difference between the available chelates.
The incidence of nausea has been reported to be 1.5% to 3.2%, and
the incidence of hives has been reported as 0.3% to <2%.
More serious reactions to IV gadolinium are very uncommon.
Postmarketing surveillance after >5 million administrations of
gadopentetate dimeglumine (Gd-DTPA) reported 15 cases of edema of
the epiglottis, 13 cases of anaphylactic shock, and only 1 death
Prior anaphylactic reaction to iodinated contrast carries an
increased risk of reaction to gadolinium, and patients with such a
history should be carefully monitored.
Gadolinium contrast media is non-nephrotoxic when given
intraven-ously, even at the higher dosages likely to be used with
DSA. A study of 64 patients, many with underlying renal
insufficiency, demonstrated a mean decrease in creatinine of 0.07
mg/dL following administration of 0.2 to 0.4 mmol/kg gadolinium.
No cases of contrast-induced nephrop-athy were reported. The same
64 patients also received iodinated contrast (typically 30 to 60 g
iodine) with a mean increase in creatinine of +0.35 mg/dL and 11
cases of contrast-induced renal failure with creatinine elevation
of 0.5 mg/dL or more.
Less information is available to assess the safety of
intra-arterial injection of gadolinium, and, again, it should be
stressed that this represents an off-label use. At the full
strength required for DSA, all available gadolinium media are
hypertonic with respect to plasma, although some preparations are
less so (Table 1). It has been suggested that the lower osmolality
formulas should be used for intra-arterial studies. Extremity pain
during peripheral injection is known to occur more commonly with
the higher osmolality formulas.
Hyperosmolality is also a concern when injecting directly into
the renal arteries since this is a potential cause of
nephrotoxicity. A study of 20 patients undergoing arteriography
with gadodiamide found that 40% had a post-procedural creatinine
elevation of greater than 0.5 mg/dL.
However, the doses used in that study were very high, up to 2.9
mmol/kg, far above the dose used in any standard MRI application.
In a rat model, gadoterate meglumine (Gd-DOTA) at 1300 mOsm/kg
infused into the aorta caused no change in creatinine or creatinine
On the other hand, the same study showed Gd-DTPA at 1900 mOsm/kg to
cause a mean 50% rise in creatinine at 24 hours post-procedure.
There are a few case reports of worsening renal function in humans
after receiving intra-arterial gadolinium.
While the intra-arterial usage of gadolinium appears to be safe, it
remains prudent to limit the total dose.
An important point has been made, however, questioning whether a
small dose of iodinated contrast is just as safe as gadolinium with
regard to nephrotoxicity.
It should be kept in mind that a standard concentration of 0.5
mmol/mL gadolinium is equi-attenuating to a comparatively dilute
concentration of 60 to 80 mg/mL of iodinated contrast. Commercially
available iodinated contrast media can be mixed with saline to
achieve a solution that is isotonic to plasma and still provides
this concentration of iodine. Furthermore, if 0.3 mmol/kg/body
weight of gadolinium contrast is accepted as a reasonable dosage
limit, then the equi-attenuating dose of iodine would be only
approximately 3 g of total iodine for an average-sized patient. (By
comparison, a typical contrast-enhanced computed tomography
examination uses about 30 g of iodine.) It has been argued that
such a low dose of iodine delivered as an isotonic solution does
not pose a significant risk of renal injury.
This discussion applies only to the potential risk of
nephrotoxicity, however. Allergic reactions to contrast media are
idiosyncratic and not dose related.
For patients with pre-existing renal failure, gadolinium is
readily removed by hemodialysis. Intra-arterial gadolinium has also
been used safely for evaluation of the carotid arteries, with no
adverse neurologic reactions reported in one series of 12 patients.
Rewindowing of gadolinium images to optimize image contrast
often makes them appear grainy due to quantum mottle (Figure 2).
While these images may be less aesthetically pleasing, diagnostic
quality is roughly equal to iodine images for medium- and
large-sized vessels. A study of the renal artery in a rat model
showed no difference in accuracy of stenosis evaluation between
gadolinium and iodine.
The same study demonstrated gadolinium to be superior to CO
imaging. A human study demonstrated little difference in diagnostic
confidence between the two contrast media for evaluation of the
aorta and main renal arteries. Evaluation of intrarenal vasculature
was poorer with gadolinium, however.
An example of a typical renal angiogram using gadolinium is shown
in Figure 3.
A series of 12 patients receiving both gadolinium and iodinated
contrast media for evaluation of the carotid artery demonstrated
slightly poorer quality images with gadolinium, but there was no
statistically significant difference in measurement of degree of
Good results have also been obtained using gadolinium for
evaluation of hemodialysis fistulas.
One postulated drawback of gadolinium imaging is that high-grade
stenoses may be misinterpreted as complete stenoses due to the
possible poor visualization of a very small patent luminal
Similarly, gadolinium may not be a good candidate for bleeding
studies, as the minimum bleeding rate that could be seen using
gadolinium is not known.
Gadolinium contrast media in conjunction with DSA is a safe and
effective imaging option for patients with a contraindication to
the use of iodinated contrast, although this is an off-label use of
gadolinium. Image quality is optimized by using a 90 to 100 kVp
beam. All available chelates contain the same concentration of
gadolinium and, therefore, provide equal vascular opacification.
However, lower osmolality preparations cause less pain when
injected into the extremities, and they may also have lower
potential for nephrotoxicity when injected into the renal
All images are courtesy of Daniel Brown, MD, Mallinckrodt
Institute of Radiology, St. Louis, MO.