Spectral Adiabatic Inversion Recovery
(SPAIR) MR imaging of the Abdomen
Thomas C. Lauenstein
Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Germany
Magnetic resonance imaging (MRI) has form in abdominal imaging is the use of fat spins have zero longitudinal magneti-
become a major imaging tool for the a 180° excitation pre-pulse, which sup- zation at this point they will not contrib-
depiction and characterization of abdom- presses the signal specific tissue de- ute to the MR signal.
inal disease. Standard abdominal MRI pending on the inversion time applied.
protocols encompass different forms of The inversion time (TI) is set according Clinical applications
T1-weighted (T1w) and T2-weighted to the T1 of fat in order to selectively Homogeneity and
(T2w) data acquisition. These sequences null the fat signal (TI = 150-170 ms). In degree of fat suppression
can be collected in less than 20 seconds, the most common implementation, the The implementation of SPAIR fat sup-
which typically is within the patients’ inversion pulse is applied with a wide pression techniques will result in a more
ability to suspend respiration. Hence, ar- frequency bandwidth to include both fat profound and homogenous fat satura-
tifacts due to physiological motion in- and water spins. A potential drawback to tion compared to conventional fat sup-
cluding respiration and bowel motion can this approach is that the water signal pression techniques. In a recent study,
be reduced, if not avoided. While most will not be fully recovered during data SNR of mesenteric and retroperitoneal
T1-weighted imaging techniques of the acquisition, and the overall water signal- fat was measured for both IR and SPAIR
abdomen include gradient echo (GRE) to-noise ratio (SNR) will be diminished. fat suppression in conjunction with
sequences, T2-weighted imaging is based This can negatively impact the contrast- T2-weighted SSFSE imaging in order to
on the collection of single shot fast spin to-noise ratio (CNR) of lesions surrounded determine the degree of fat suppression
echo (SSFSE) data. The latter sequences by tissue, such as tumors within the liver. . The study showed that improved
in conjunction with fat saturation play fat suppression was found when SPAIR-
a key role for the interpretation of differ- Technical considerations SSFSE was applied (Fig. 1).
ent abdominal processes as liver lesions for SPAIR
can be most accurately delineated and The inversion recovery (IR) technique Depiction of anatomical structures
specified . Furthermore, T2-weighted can be modified by using chemical selec- An advantage of SPAIR compared to con-
imaging with fat saturation is crucial tive or spectral pre-saturation attenuat- ventional IR techniques is demonstrated
for the depiction of edema and/or free ed inversion-recovery pre-pulses. SPAIR by the improvement in CNR of the he-
fluid. This is particularly helpful for the (Spectral Adiabatic Inversion Recovery) patic lesions. The better liver lesion con-
depiction of inflammatory processes is a powerful technique for fat suppres- trast on SPAIR-SSFSE images is consis-
of the bowel, e.g. in patients with sion which offers different advantages tent with the predicted benefits of
Crohn’s disease [2, 3], appendicitis [4, 5] over conventional fat suppression tech- applying a frequency-sensitive inversion
or diverticulitis [6–8]. Finally, T2-weight- niques. The technique is insensitive to pulse. This leaves the maximum possible
ed data may be particularly useful in B1 inhomogenities and only fat spins are water signal intact as only the fat spins
the setting of pregnant patients*. As the suppressed/inverted. SPAIR uses a spec- are inverted. Two types of focal liver
intravenous administration of gadolini- trally selective adiabatic inversion pulse lesions have been evaluated : heman-
um based contrast agents is contraindi- to invert the fat spins in the imaging vol- giomas with a relatively high CNR and
cated in this patient group, T1-weighted ume. After the adiabatic pulse a large metastases with a relatively low CNR.
imaging is restricted and only provides spoiler is utilized in order to destroy any The CNR was found significantly increased
limited information. Hence, T2-weighted transverse magnetization. The fat spins for both families of lesions when using
imaging with fat saturation has been will now decay according to the T1 re- SPAIR compared to IR SSFSE (Figs. 2
found to be the key sequence in preg- laxation rate and after a certain charac- and 3).
nant* women with suspected abdominal teristic time (TI null) the longitudinal Furthermore, delineation of bowel
inflammation or tumor disease [9–12]. magnetization will be zero. At this time wall structures is markedly improved on
Different techniques for fat saturation point the excitation pulse of the SSFSE SPAIR SSFSE (Fig. 4). This improvement
in MRI can be used. The most common T2-weighted module is applied. As the is due to two different factors that dif-
16 MAGNETOM Flash · 2/2008 · www.siemens.com/magnetom-world
1 Homogeneous fat saturation in the retroperitoneum (dashed 2 Patient with liver metastases (arrow) of colorectal cancer.
arrow) and the mesenteries (arrow) with the SPAIR technique. The lesion is evident and provides high CNR values on SPAIR
3 Patient with several hemangiomas (arrows). 4 Conspicuous bowel loops (arrow) using the SPAIR technique.
SPAIR T2-weighted MRI.
MAGNETOM Flash · 2/2008 · www.siemens.com/magnetom-world 17
5 Patient with active colitis. There
is increased contrast enhancement
after iv gadolinium administration
shown on T1-weighted GRE imag-
ing (5A; arrow). A high T2 signal of
the bowel wall can be depicted on
T2-weighted SPAIR images (5B;
arrow), which is consistent with ac-
tive inflammatory disease due to
6 Patient with mildly active in-
flammatory changes of the as-
cending colon (arrow). T1-weight-
ed contrast-enhanced MRI reveals
increased contrast uptake of the
inflamed bowel segment and
thickening of the bowel wall (6A).
The T2 signal on the SPAIR image
is only slightly elevated (6B).
7 Patient with non-active / fibrot-
ic inflammation of the sigmoid co-
lon (arrow). Similar to the active
forms of inflammatory bowel dis-
ease (IBD) there is increased con-
trast enhancement on T1-weight-
ed MRI (7A). However, there is
lack of edema, and thus the T2
signal is not elevated on the SPAIR
18 MAGNETOM Flash · 2/2008 · www.siemens.com/magnetom-world
8 SPAIR T2-weighted SSFSE MRI can be used as a stand-alone sequence for therapeutic monitoring. This patient presented with sign of active
inflammation in the terminal ileum and highly elevated T2 signal on SPAIR imaging (8A). One week after the initiation of anti-inflammatory
medication the T2 signal dropped as a correlation of therapeutic response (8B).
ferentiate SPAIR SSFSE: one factor is the were either time-consuming (e.g. MR flammation show an increased contrast
relatively greater sensitivity to motion based perfusion analyses), invasive enhancement . Hyperintensity on
of standard IR SSFSE. In addition, bowel (colonoscopy / biopsy) or inaccurate T2-weighted images, however, is related
wall visualization should benefit from (CDAI). Hence, a relatively fast, simple to increased edema and inflammatory
the increased SNR of water-containing and non-invasive technique is desired fluid components within or adjacent to
structures on SPAIR SSFSE. in appraising the level of inflammatory the bowel wall, whereas T1-weighted
activity and also in following up these hyperintensity may be attributed to a
Inflammatory abdominal processes patients for treatment response. hypervascularity (in active disease) or a
Evaluation of disease activity in patients SPAIR T2-weighted SSFSE sequences and delayed wash-out (in fibrotic /chronic
with inflammatory bowel disease (IBD) gadolinium enhanced T1-weighted se- disease).
is often a challenging clinical situation. quences are complementary techniques Examples of contrast-enhanced
While active inflammation is treated in patients with IBD . Gadolinium- T1-weighted GRE images and SPAIR
with systemic corticosteroids or other enhanced T1-weighted data is helpful T2-weighted SSFSE images are shown in
immuno-modulator drugs, surgical ther- to detect IBD independent of its activity figures 5–7 for highly active, intermedi-
apeutic options are chosen for chronic state with a high sensitivity. However, ate active and non-active IBD. Once the
disease. This discrepancy in therapy accuracy of T1-weighted imaging to diagnosis of IBD is established, SPAIR
strategies underlines the need for an differentiate between active and non- T2-weighted SSFSE imaging can be used
accurate categorization and differentia- active disease is only moderate. En- as a stand-alone sequence for therapy
tion between active and chronic disease. hancement patterns of T1-weighted im- monitoring (Fig. 8). Furthermore, this
Attempts of classifying IBD in the past aging are unspecific: both bowel method is also very helpful not only for
were based on different variables that segments with active and chronic in- the assessment of IBD including Crohn’s
* The safety of imaging fetuses/infants has not been established.
MAGNETOM Flash · 2/2008 · www.siemens.com/magnetom-world 19
disease and Ulcerative colitis, but also 9
9 Patient with
for diverticulitis (Fig. 9) and the depic-
tion of fistulae (Fig. 10). There is increased
Conclusion in and adjacent to
the wall of the sig-
There are overall benefits of SPAIR SSFSE
moid colon (arrow)
that can be measured on clinical abdom- in keeping due to
inal MR images regarding fat saturation, edema.
particularly in fat adjacent to bowel and
for improving overall image contrast
even between non-fatty soft tissues, such
as can be demonstrated with liver mass-
es. Furthermore, SPAIR SSFSE is a crucial
tool for the depiction of inflammatory
processes in the abdomen, particularly 10
IBD. By means of SPAIR T2-weighted 10 T2-weighted
SPAIR imaging can
SSFSE a differentiation between active
easily display not
and non-active inflammatory processes only an inflamma-
can be easily established. tory process itself,
but also complica-
tions such as a fluid-
Contact filled fistula – be-
Thomas C. Lauenstein, M.D. tween bowel and
University Hospital Essen cutis – (arrow).
Dept. of Diagnostic and Interventional
Radiology and Neuroradiology
45122 Essen, Germany
1 Gaa J, Hatabu H, Jenkins RL, Finn JP, Edelman RR. Abdom Imaging 2003;28:794–798. Tudorascu D, Martin DR. Evaluation of optimized
Liver masses: replacement of conventional 7 Heverhagen JT, Ishaque N, Zielke A, et al. Feasi- inversion-recovery fat-suppression techniques
T2-weighted spin-echo MR imaging with breath- bility of MRI in the diagnosis of acute diverticu- for T2-weighted abdominal MR Imaging. J Magn
hold MR imaging. Radiology 1996;200: litis: initial results. Magma 2001;12:4–9. Reson Imaging 2008:27:1448–1454.
459–464. 8 Heverhagen JT, Zielke A, Ishaque N, Bohrer T, 14 Maccioni F, Bruni A, Viscido A, et al. MR imag-
2 Florie J, Wasser MN, Arts-Cieslik K, Akkerman El-Sheik M, Klose KJ. Acute colonic diverticulitis: ing in patients with Crohn disease: value of
EM, Siersema PD, Stoker J. Dynamic contrast- visualization in magnetic resonance imaging. T2- versus T1-weighted gadolinium-enhanced
enhanced MRI of the bowel wall for assessment Magn Reson Imaging 2001;19:1275–1277. MR sequences with use of an oral superpara-
of disease activity in Crohn’s disease. AJR Am J 9 Birchard KR, Brown MA, Hyslop WB, Firat Z, Se- magnetic contrast agent. Radiology
Roentgenol 2006;186:1384–1392. melka RC. MRI of acute abdominal and pelvic 2006;238:517–530.
3 Maccioni F, Bruni A, Viscido A, et al. MR imaging pain in pregnant patients. AJR Am J Roentgenol 15 Udayasankar UK, Lauenstein TC, Martin DR. Role
in patients with Crohn disease: value of T2- ver- 2005;184:452–458. of Spectral Presaturation Attenuated Inversion-
sus T1-weighted gadolinium-enhanced MR se- 10 Brown MA, Birchard KR, Semelka RC. Magnetic Recovery (SPAIR) T2 Fat-suppressed MR Imaging
quences with use of an oral superparamagnetic resonance evaluation of pregnant patients with in Active Inflammatory Bowel Disease. RSNA
contrast agent. Radiology 2006;238:517–530. acute abdominal pain. Semin Ultrasound CT MR 2007.
4 Martin DR, Danrad R, Herrmann K, Semelka RC, 2005;26:206–211.
Hussain SM. Magnetic resonance imaging of the 11 Cobben LP, Groot I, Haans L, Blickman JG, Puy-
gastrointestinal tract. Top Magn Reson Imaging laert J. MRI for clinically suspected appendicitis
2005;16:77–98. during pregnancy. AJR Am J Roentgenol
5 Nitta N, Takahashi M, Furukawa A, Murata K, 2004;183:671–675.
Mori M, Fukushima M. MR imaging of the normal 12 Tang Y, Yamashita Y, Takahashi M. Ultrafast T2-
appendix and acute appendicitis. J Magn Reson weighted imaging of the abdomen and pelvis:
Imaging 2005;21:156–165. use of single shot fast spin-echo imaging. J
6 Cobben LP, Groot I, Blickman JG, Puylaert JB. Magn Reson Imaging 1998;8:384–390.
Right colonic diverticulitis: MR appearance. 13 Lauenstein TC, Sharma P, Hughed T, Heberlein K,
20 MAGNETOM Flash · 2/2008 · www.siemens.com/magnetom-world