Biomarkers and Genomic Medicine (2013) 5, 39e43

Available online at www.sciencedirect.com

journal homepage: www.j-bgm.com

ORIGINAL ARTICLE

Correlation between SPGR MR signal intensity and

iron concentration: Phantom study
Ching-Hui Yang a, San-He Hong a,y, Jo-Chi Jao b, Shu-Neng Tsai c,
Chia-Chi Shiao d, Kwok-Wan Yeung a,**, Po-Chou Chen c,*

a
Department of Radiology, Fooyin University Hospital, Pingtung, Taiwan
b
Department of Medical Imaging and Radiological Sciences, College of Health Sciences, Kaohsiung Medical University,
Kaohsiung, Taiwan
c
Department of Biomedical Engineering, I-Shou University, Kaohsiung, Taiwan
d
Department of Radiology, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan

Received 18 March 2013; received in revised form 17 April 2013; accepted 23 April 2013
Available online 29 May 2013

KEYWORDS Abstract Brain iron deposits have been demonstrated to be an important indicator in the
echo time; pathophysiology of neurodegenerative processes. Hence, it is crucial to study the basic char-
flip angle; acteristics of ferric ion by using a noninvasive imaging modality, such as magnetic resonance
magnetic resonance imaging (MRI). Spoiled gradient recalled echo (SPGR) is one of the rapid MRI scanning pulse se-
imaging; quences and has been widely used in clinical diagnosis. There are three important scanning pa-
repetition time; rameters, i.e., repetition time (TR), echo time (TE), and flip angle (FA), which will affect the
spoiled gradient SPGR MR signal intensity (SI) and contrast of investigated tissues and organs. The purpose of
recalled echo this study is to investigate the variability of SPGR MR SIs of ferric iron solutions with various
concentrations by changing the TR, TE, and FA using a 1.5-T clinical whole body scanner. It
was found that the optimal flip angle, i.e., the Ernst angle, increased with the concentration
of ferric ions. The SI increased rapidly at smaller TRs and then rose slowly as the TR became
larger. The SI was also found to be smaller for longer TE. Furthermore, there was better line-
arity between the MR, SI, and ferric iron concentration for the smaller TE, smaller TR, and
larger FA.
Copyright ª 2013, Taiwan Genomic Medicine and Biomarker Society. Published by Elsevier
Taiwan LLC. All rights reserved.

* Corresponding author. No. 8, Yida Road, Yanchao District, Kaohsiung City 82445, Taiwan.
** Corresponding author. Fooyin University Hospital, Number 5, Jhongshan Road, Donggang, Township, Pingtung, Taiwan.
E-mail addresses: 9769@fy.org.tw (K.-W. Yeung), pcchen@isu.edu.tw (P.-C. Chen).
y
San-He Hong contributed equally with Ching-Hui Yang to this work.

2211-4254/$36 Copyright ª 2013, Taiwan Genomic Medicine and Biomarker Society. Published by Elsevier Taiwan LLC. All rights reserved.
http://dx.doi.org/10.1016/j.gmbhs.2013.04.005
40
Introduction
Brain iron deposits have recently been reviewed and
demonstrated to be an important indicator in pathophysi-
ology or neurodegenerative diseases.1e5 The conventional
assessments of ferric iron are done in invasive ways such as
inductively coupled plasma mass spectrometry. Recently,
brain iron quantification using noninvasive magnetic reso-
nance imaging (MRI) has drawn the attention of researchers.6
Alzheimer’s and Parkinson’s diseases have been found to have
increased iron levels in the corresponding regions of the
human brain.7e9 Similar findings have also been reported in
multiple sclerosis and brain ageing.10 The accurate quantifi-
cation of iron content in the human brain may well lead to a
better understanding of the disease progression. There are
various forms of iron in human brain, and how to quantify
them separately and accurately will be a major challenge.
Since the correlation between MRI characteristics and ferric
ion concentrations is not well understood, it is crucial to study
the basic MRI characteristics of ferric ions by using a phantom
containing ferric iron solutions with various concentrations.
Spoiled gradient recalled echo (SPGR) is one of the rapid
scanning pulse sequences in MRI11e14 and has been widely used
in clinical routine examinations nowadays.15e23 The SPGR
pulse sequence randomly changes the phase of radiofrequency
(RF) pulse to spoil the transverse magnetization and the
resulting image contrast is mainly T1- or PD-weighted imaging.
The SPGR MR signal intensity (SI) of investigated tissues and
organs is a function of three important scanning parameters,
i.e., repetition time (TR), echo time (TE), and flip angle (FA). It
has been reported that the tissue contrast is very sensitive to
the flip angle in SPGR MRI.24 The aim of this study was to
investigate the variability of the SPGR MR SI of a phantom
containing ferric iron solutions with various concentrations by
changing the scanning parameters TR, TE, and FA. To have an
appropriate choice of MR scanning parameters can demon-
strate better image contrast, and it enables us to differentiate
abnormal tissues from normal tissues, which will be helpful for
the early and accurate diagnosis of iron deposition.
Materials and methods
Phantom preparation
The experimental phantom contained six test tubes with
different concentrations of ferric iron in distilled water
Figure 1 Procedures for formulating the ferric iron solution.
C.-H. Yang et al.
solutions (30 mL). The in vivo ferric iron content of a 70-kg
normal adult has been reported to be about 5 g.25,26 The
average density of adults is about 1.00084 g/cm3; hence,
the ferric iron concentration corresponds to 1.28 mM. The
ferric iron compositions in human body are commonly Fe3þ
or Fe2þ.27e29 Hence, distilled water solution of FeCl3 was
used in this study to simulate the ferric iron concentration
in human body. The formulations of Fe3þ solutions were
0 mM, 0.25 mM, 0.5 mM, 0.75 mM, 1.0 mM, and 1.25 mM.
Hydrolysis is always accompanied with the ferric ion solu-
tion, which results in the sediment of the ferric compound.
In order to avoid the sediment of the ferric compound, 1%
of HCl (99% HCl, Sigma-Aldrich, St. Louis, MO, USA) was
added into each test tube. The flow chart of formulating
the ferric iron solution is shown in Fig. 1. Using a volumetric
cylinder (100 mL) to measure 99 mL of double-deionized
water, 1 mL HCl (99% HCl, Sigma-Aldrich, St. Louis, MO, US),
and various amounts of FeCl3 (iron chloride, anhydrous,
98%, Alfa Aesar, A Johnson Matthey Co, Lancashire, UK)
were mixed into a 500-mL beaker. Then, the beaker was
placed on a Corning PC-410D magnetic stirrer with heating
(Corning Inc, Tewksbury, MA, USA) by stirring until a ho-
mogeneous solution was obtained after a stir bar (Teflon,
10  350 mm, Shimada Instrument CO., LTD., Tokyo, Japan)
was immersed into it. The test tubes were arranged to form
a triangle on a styrofoam holder as shown in Fig. 2. After-
wards, the phantom was put into a head coil for scanning.
MRI study
All experiments were performed using a GE Signa HDxt 1.5 T
whole body magnetic resonance scanner (GE healthcare,
Milwaukee, WI, USA) equipped with an eight-channel high
resolution head coil. After a three-plane localizer scan, a
SPGR pulse sequence was performed at various TRs, TEs, and
FAs. The FA varied from 5 to 90 with a 5 increment when
TR Z 20 ms and TE Z 7.1 ms. The TR varied from 20 ms to
300 ms with a 20-ms increment when TE Z 3.1 ms and
FA Z 90 . TE was varied from 5 ms to 40 ms with 5-ms in-
crements when TR Z 50 ms and FA Z 90 . Afterwards, MR
images acquired from SPGR scanning were retrieved to
Advantage Window (AW, GE healthcare, Milwaukee, WI, USA)
workstation to measure the region-of-interest (ROI) values
for each test tube. The ROI values of ferric solutions versus
TR, TE, and FA values at various concentrations as well as
those versus ferric ion concentrations at various TR, TE, and
MR signal intensity and iron concentration
0.25 mM 0.5 mM 0.75 mM
1.0 mM 1.25 mM
0 mM
Figure 2 The experimental phantom setup.
FA values were plotted. The statistical analysis of the SI
versus iron concentration with various FAs, TEs, and TRs was
performed by the nonparametric Friedman test using com-
mercial SPSS version 12.0 (SPSS Inc., Chicago, IL, USA). A p-
value < 0.05, was considered to have significance.
Results
Fig. 3 shows a representative SPGR MR image. The curves of
the SPGR SI versus repetition time for different ferric ion
concentrations are shown in Fig. 4A. The SI increased
rapidly at a small TR and then rose slowly as the TR became
larger for all different ferric ion concentrations. The curves
of SPGR SI versus ferric ion concentration at various TRs are
shown in Fig. 4B. There was better linearity between the
MR SI and ferric ion concentration when the TR was smaller.
The curves of the SPGR SI versus FA at various ferric ion
concentrations are shown in Fig. 5A. The optimal flip angle,
i.e., the Ernst angle, increased with the concentration of
Fe3þ. The curves of SPGR SI versus ferric ion concentration
Figure 3 A representative SPGR MR image (FA Z 90 ,
TR Z 40 ms and TE Z 3.1 ms). FA Z flip angle; MR Z magnetic
resonance; SPGR Z spoiled gradient recalled echo; TE Z echo
time; TR Z repetition time.
41
A
B
Figure 4 (A) SPGR SI versus TRs at various ferric ion con-
centrations, (B) SPGR SI versus ferric ion concentration at
various TRs (p < 0.05). SI Z signal intensity; SPGR Z spoiled
gradient recalled echo; TR Z repetition time.
at various FAs are shown in Fig. 5B. There was better
linearity between the MR SI and ferric ion concentration
when the FA was larger. The curves of the SPGR SI versus TE
at various ferric ion concentrations are shown in Fig. 6A.
The SI decreased with TE. Furthermore, the larger the
concentration of Fe3þ, the higher was the SPGR SI. The
curves of the SPGR SI versus ferric ion concentration at
various TEs are shown in Fig. 6B. There was better linearity
between the MR SI and ferric ion concentration when the TE
was smaller. There were significant differences in the SI
versus iron concentration among various TRs (Fig. 4), as
well as FAs (Fig. 5) and TEs (Fig. 6) using the Friedman test
(p < 0.05).
Discussions
Brain iron deposits have been known as an important factor
related to brain degeneration. MRI is noninvasive and has no
ionizing radiation. Therefore, it is a good modality to
investigate brain iron deposits. The SPGR pulse sequence is
commonly used for fast MRI. This study investigated the
variability of SPGR MRI SI of a phantom with various Fe3þ
concentrations by changing the important scanning pa-
rameters, including TR, FA, and TE.
42
A A
B B
Figure 5 (A) SPGR SI versus FAs at various ferric ion con-
centrations, (B) SPGR SI versus ferric ion concentration at
various FAs (p < 0.05). FA Z flip angle, SI Z signal intensity;
SPGR Z spoiled gradient recalled echo.
The SI increases along with the increasing ferric ion
concentration for all flip angles. The result implies that the
larger the ferric ion concentration, the shorter is the T1
relaxation time of the ferric ion solution. The recovery of
spins becomes faster; hence, the SI increases as the ferric
ion concentration increases. It is known that the SI in-
creases rapidly at a smaller flip angle from the individual
ferric ion curve. The increase in SI becomes slower as the
flip angle becomes larger. Afterwards, the SI gradually de-
creases along with a larger flip angle. As the SI reaches the
maximum value, the flip angle is the so-called optimal flip
angle (i.e., the Ernst angle). The optimal flip angle in-
creases with the concentration of ferric ions.
To have better linearity between the MR signal and ferric
ion concentration, it is suggested to choose a larger FA.
The SI increases rapidly at small TR and then the varia-
tions gradually become smaller as TR becomes larger for all
different ferric ion concentrations. It is also found that the
variation in SI is smaller for larger ferric ion concentrations.
To have a better linearity between the MR SI and ferric ion
concentration, it is suggested to choose a smaller TR when
the signal-to-noise ratio (SNR) is enough.
The higher the ferric ion concentration, the larger the SI
of SPGR is for any echo time. The SI decays faster for a
higher ferric ion concentration at a longer echo time. It
C.-H. Yang et al.
Figure 6 (A) SPGR SI versus TEs at various ferric ion con-
centrations, (B) SPGR SI versus ferric ion concentration at
various TEs (p < 0.05). SI Z signal intensity; SPGR Z spoiled
gradient recalled echo; TE Z echo time.
implies that the higher the ferric ion concentration, the
shorter is the T2 relaxation time, which results in the faster
decay of MR SI. To have a better linearity between the MR
signal and ferric ion concentration, it is suggested to choose
a smaller TE.
The SI as a function of the iron concentration [C] can be
obtained by linear regression, e.g., SI Z 548 þ 134 [C] with
R2 Z 0.9785 at TR Z 20 ms, TE Z 3.1 ms, and FA Z 90 . The
forms of iron inside the tissues, such as ferritin, transferrin,
and heme iron, are distributed nonuniformly, and the struc-
tures of tissues are more complex than the ferric ion solution
phantom. Owing to the differences between the tissues and
the phantom, the iron concentration of tissues may not be
directly calculated by this linear equation from the measured
SI. However, the optimized scanning parameters obtained in
this study can be easily applied to find the linear function
between the SI and iron concentration of tissues in vivo.
Several MRI characteristics, such as relaxation rates (R1,
R2*) and phase, have been investigated for iron quantification
in brain, liver, and myocardium.30e32 Among them, R1 is less
commonly used than R2* and phase because of its long
acquisition time. In this study, we investigated the correlation
between MR SI and iron concentration instead of relaxation
rates and phase. As the paramagnetic effect of iron, R1, and
R2* increases with iron concentration so that the SI of the T1-
MR signal intensity and iron concentration
weighted (T1W) image increases and the SI of the T2*-
weighted (T2*W) image decreases. We found that the SI of the
T1-weighted image (shorter TR, shorter TE, and larger FA) has
better linearity than the SI of the T2*-weighted image (longer
TE) within the iron concentration range of 0 e 1.25 mM. This is
because the SI of the T2*W image decreases quickly and loses
its linearity when the iron concentration increases. T1W im-
aging (T1WI) is performed routinely to obtain the anatomical
structures of tissues. When the scanning parameters are
optimized, the SI difference of the T1W image can demon-
strate the difference in relative iron concentration with a
shorter scanning time than R1.
It is important to obtain the MR images with high quality
by using optimized scanning parameters. In conclusion, this
study can provide useful information for the choice of
scanning parameters TR, FA, and TE to obtain high-quality
ferric MR images and the linearity between the SI and the
iron concentration. It can be extended to compare and
correlate all types of gradient echo pulse sequences in
future researches.
Acknowledgments
This study was supported by Fooyin University Hospital,
Pingtung, Taiwan, R.O.C., under the grant number FH-HR-
100-04.
References
1. Benarroch EE. Brain iron homeostasis and neurodegenerative
disease. Neurology. 2009;72:1436e1440.
2. Zecca L, Youdim MB, Riederer P, et al. Iron, brain ageing and
neurodegenerative disorders. Nat Rev Neurosci. 2004;5:
863e873.
3. Drysdale JW. Human ferritin gene expression. Prog Nucleic
Acid Res Mol Biol. 1988;35:127e172.
4. Harrison PM, Arosio P. The ferritins: molecular properties, iron
storage function and cellular regulation. Biochim Biophys Acta.
1996;1275:161e203.
5. Harrison PM, Treffry A, Lilley TH. Ferritin as an iron storage
protein: mechanisms of iron uptake. J Inorg Biochem. 1986;27:
287e293.
6. Martin WR. Quantitative estimation of regional brain iron with
magnetic resonance imaging. Parkinsonism Relat Disord. 2009;
15(suppl 3):S215eS218.
7. Haacke EM, Cheng NY, House MJ, et al. Imaging iron stores in
the brain using magnetic resonance imaging. Magn Reson Im-
aging. 2005;23:1e25.
8. Stankiewicz J, Panter SS, Neema M, et al. Iron in chronic brain
disorders: imaging and neurotherapeutic implications. Neuro-
therapeutics. 2007;4:371e386.
9. Zhu WZ, Zhong WD, Wang W, et al. Quantitative MR phase-
corrected imaging to investigate increased brain iron deposi-
tion of patients with Alzheimer disease. Radiology. 2009;253:
497e504.
10. Hallgren B, Sourander P. The effect of age on the non-haemin
iron in the human brain. J Neurochem. 1958;3:41e51.
11. Shogry ME, Elster AD. Cerebrovascular enhancement in spoiled
GRASS (SPGR) images: comparison with spin-echo technique. J
Comput Assist Tomogr. 1992;16:48e53.
12. Hathout G, Jamshidi N. Parameter optimization for quantita-
tive signal-concentration mapping using spoiled gradient echo
MRI. Radiol Res Pract. 2012;2012:815729.
43
13. Murase K. Generalized equation for describing the magneti-
zation in spoiled gradient-echo imaging. Magn Reson Imaging.
2011;29:723e730.
14. Haase A, Frahm J, Mattaei D, et al. FLASH Imaging: rapid NMR
imaging using low flip-angle pulses. 1986. J Magn Reson. 2011;
213:533e541.
15. Gaeta M, Scribano E, Mileto A, et al. Muscle fat fraction in
neuromuscular disorders: dual-echo dual-flip-angle spoiled
gradient-recalled MR imaging technique for quantificationea
feasibility study. Radiology. 2011;259:487e494.
16. Mike A, Glanz BI, Hildenbrand P, et al. Identification and
clinical impact of multiple sclerosis cortical lesions as assessed
by routine 3T MR imaging. AJNR Am J Neuroradiol. 2011;32:
515e521.
17. Blankenbaker DG, Ullrick SR, Kijowski R, et al. MR arthrography
of the hip: comparison of IDEAL-SPGR volume sequence to
standard MR sequences in the detection and grading of carti-
lage lesions. Radiology. 2011;261:863e871.
18. Ikawa F, Uozumi T, Kiya K, et al. Cavernous sinus meningioma
presenting as orbital apex syndrome. Diagnostic methods of
dynamic MRI, spoiled GRASS (SPGR) image. Neurosurg Rev.
1995;18:277e280.
19. Low RN, Sigeti JS. MR imaging of peritoneal disease: compar-
ison of contrast-enhanced fast multiplanar spoiled gradient-
recalled and spin-echo imaging. AJR Am J Roentgenol. 1994;
163:1131e1140.
20. Watanabe Y, Nishimura T. Brain functional MRI using 3D-
SPGRpulse sequence. Nihon Rinsho. 1997;55:1675e1678.
21. Hong HS, Yi BH, Cha JG, et al. Enhancement pattern of the
normal facial nerve at 3.0 T temporal MRI. Br J Radiol. 2010;
83:118e121.
22. Xia Y, Zheng S. Reversed laminar appearance of articular
cartilage by T1-weighting in 3D fat-suppressed spoiled gradient
recalled echo (SPGR) imaging. J Magn Reson Imaging. 2010;32:
733e737.
23. Studler U, White LM, Andreisek G, et al. Impact of motion on
T1 mapping acquired with inversion recovery fast spin echo
and rapid spoiled gradient recalled-echo pulse sequences for
delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) in
volunteers. J Magn Reson Imaging. 2010;32:394e398.
24. Reeder SB, McVeigh ER. Tag contrast in breath-hold CINE car-
diac MRI. Magn Reson Med. 1994;31:521e525.
25. Vymazal J, Righini A, Brooks RA, et al. T1 and T2 in the brain of
healthy subjects, patients with Parkinson disease, and patients
with multiple system atrophy: relation to iron content. Radi-
ology. 1999;211:489e495.
26. Chen JC, Hardy PA, Kucharczyk W, et al. MR of human post-
mortem brain tissue: correlative study between T2 and assays
of iron and ferritin in Parkinson and Huntington disease. AJNR
Am J Neuroradiol. 1993;14:275e281.
27. Bartzokis G, Mintz J, Sultzer D, et al. In vivo MR evaluation of
age-related increases in brain iron. AJNR Am J Neuroradiol.
1994;15:1129e1138.
28. Bartzokis G, Cummings JL, Markham CH, et al. MRI evaluation
of brain iron in earlier- and later-onset Parkinson’s disease and
normal subjects. Magn Reson Imaging. 1999;17:213e222.
29. Ordidge RJ, Gorell JM, Deniau JC, et al. Assessment of relative
brain iron concentrations using T2-weighted and T2*-weighted
MRI at 3 Tesla. Magn Reson Med. 1994;32:335e341.
30. Ropele S, de Graaf W, Khalil M, et al. MRI assessment of iron
deposition in multiple sclerosis. J Magn Reson Imaging. 2011;
34:13e21.
31. Tziomalos K, Perifanis V. Liver iron content determination by
magnetic resonance imaging. World J Gastroenterol. 2010;16:
1587e1597.
32. Cheung JS, Au WY, Ha SY, et al. Reduced transverse relaxation
rate (RR2) for improved sensitivity in monitoring myocardial iron
in thalassemia. J Magn Reson Imaging. 2011;33:1510e1516.