Transcript Update on Imaging

Update on Imaging: Detection
of Iron in Liver and Heart
Tim St. Pierre, BSc, PhD
Professor
School of Physics
The University of Western Australia
Crawley, Australia
Iron Loading Is Different
in Different Organs
Why Is Measurement of Liver Iron
Concentration Important?
• A patient’s liver iron concentration (LIC) value is
the best measure of total body iron stores
• A patient’s LIC value enables better informed
decisions on when to
–
–
–
–
Initiate chelation therapy
Increase chelation dose
Decrease chelation dose
Change mode of chelator delivery (eg, IV mode)
LIC Is a Reliable Measure of Total
Body Iron Stores in Patients with
Thalassaemia Major
There is a very strong
correlation between LIC
and total body iron
stores in thalassaemia
major patients
Abbreviation: LIC, liver iron concentration.
With permission from Angelucci E, et al. N Eng J Med. 2000;343:327-331.
LIC Thresholds and Associated Risks
LIC Threshold Clinical Relevance
(mg Fe/g dry
weight)
1.8
3.2
Upper 95% of normal
7.0
Suggested upper limit of optimal range for LICs for
transfusional iron overload and threshold for
increased risk of iron-induced complications1
15.0
Suggested lower limit of optimal range for LICs for
chelation therapy in transfusional iron overload1
Threshold for greatly increased risk for cardiac
disease and early death in patients with transfusional
iron overload1
1. Olivieri NF, Brittenham GM. Blood. 1997;89:739-761.
LIC and Long-Term Prognosis
Initial LIC
(mg Fe/g dw)
13-Year Cardiac
Disease–Free
Survival
Patients (n)
Group
<7
93.3% (SE 6.4)
15
(i)
7–15
71.4% (SE 17.1)
7
(ii)
>15
50.0% (SE 15.8)
10
(iii)
32 thalassaemia major patients followed for median period of
13.6 years after single biopsy LIC measurement
Telfer PT, et al. Br J Haematol. 2000;110:971-977.
Why Is Measurement of Heart Iron
Important?
• Heart iron measurements (by cardiac MRI) have
been shown to have a high sensitivity and
specificity for predicting cardiac failure within 12
months for thalassaemia major patients
• In a study of 652 thalassaemia major patients
– 83% of patients who developed arrhythmia had a
cardiac T2* of <20 ms
– 98% of patients who developed heart failure had a
cardiac T2* of <10 ms
Kirk P, et al. Circulation. 2009;120: in press.
Relationship Between
Liver and Heart Iron
Heart Iron Changes Generally Lag
Behind Liver Iron Changes
With permission from Noetzli LJ, et al. Blood. 2008;112:2973-2978.
Strengths and Weaknesses of
Various Imaging Methods
Methods of Monitoring Iron Loading
• Serum ferritin
• Liver biopsy
• Biomagnetic susceptometry
• MRI
Methods of Monitoring
Body Iron Stores
Is Serum Ferritin a Reliable
Indicator of LIC?
• Cross-sectional study of 37 patients with sickle cell
anaemia and 74 patients with thalassaemia major
• Only 57% of the variability in plasma ferritin concentration
could be explained by the variation in hepatic iron stores
• The 95% prediction intervals for hepatic iron concentration,
given the plasma ferritin, were so broad as to make a single
determination of plasma ferritin an unreliable predictor of
body iron stores
• Eg, given a plasma ferritin of 1000 ng/mL, the 95%
prediction interval for hepatic storage iron was 0–6.948 mg
Fe/g liver, wet weight
Brittenham GM, et al. Am J Hematol. 1993;42:81-85.
Serum Ferritin in Thalassaemia
Major and Intermedia
Serum ferritin has almost no sensitivity or specificity for iron stores in
thalassaemia intermedia
Abbreviations: TI, Thalassaemia intermedia; TM, Thalassaemia major.
With permission from:
Origa R, et al. Haematologica. 2007;92:583-588.
With permission from:
Taher A, et al. Haematologica. 2008;93:1584-1585.
Serum Ferritin
• Serum ferritin can be used for monitoring trends
in patient transfusional iron loading
• Serum ferritin does not give reliable information
on degree of patient iron loading
Measuring Liver Iron Concentration
by Biopsy
• Methods
– Percutaneous
– Laparoscopic
– Transjugular
• Risk of Complications
–
–
–
–
–
Death
Bile leak
Bleeding
Any pain
Significant pain
1:10,000–1:12,000
1:1,000
1:100
1:4
1:10–1:20
Siegel CA, et al. Cleve Clin J Med. 2005;72:199-224.
Heterogeneity of Iron Concentration
Throughout the Liver
Sample Size and Type
CV of LIC
Pathology
Source
Needle biopsy
(<4 mg dw)
19%
Normal
Emond, et al. 1999
Kreeftenberg, et al. 1984
Needle biopsy
(<4 mg dw)
>40%
End-stage
liver disease
Emond, et al. 1999
Kreeftenberg, et al. 1984
Needle biopsy
(9 mg dw)
9%
Normal
Barry, Sherlock. 1971
“Cubes”
(200–300 mg wet weight)
17%
24%
-thalassaemia
Ambu, et al. 1995
“Cubes”
(1000–3000 mg wet weight)
19%
-thalassaemia
Part-cirrhotic
Noncirrhotic
Clark, et al. 2003
Abbreviations: CV, coefficient of variation; dw, dry weight; LIC, liver iron concentration.
Ambu R, et al. J Hepatol.1995;23:544-549. Barry M, Sherlock S. Lancet.1971;1:100-103. Clark PR, et al.
Magn Reson Med. 2003;49:572-575. Emond MJ, et al. Clin Chem.1999;45:340-346. Kreeftenberg HG, et al.
Clin Chim Acta. 1984;144:255-262.
Noninvasive Methods of
Tissue Iron Measurement
Biomagnetic Liver Susceptometry
(SQUID)
Biomagnetic Liver Susceptometry
Liquid helium
Cryogenic package
Bellows
Liver
Liver
Fischer R. In: Magnetism In Medicine: A Handbook. Wiley-VCH;1998:286-301.
Needle Biopsy LIC vs
Biomagnetic Liver Susceptometry
• There is a good correlation between LIC
by biopsy and LIC by SQUID up to 3.5 mg
Fe/g wet tissue
• Above 3.5 mg Fe/g wet tissue, correlation
decreases, most likely because of
increased sampling error on biopsy
Fischer R. In: Magnetism In Medicine: A Handbook. Wiley-VCH;1998:286-301.
Noninvasive Methods of
Tissue Iron Measurement
Magnetic Resonance Imaging
(MRI)
Principles of MRI
• Magnetic field and radio signal pulses
• Initial pulse excites protons in tissue
• Signal received from tissue decays with time after
initial pulse
• Rate of decay different for different tissues
• Rate of decay highly influenced by presence of iron
• Rate known as either R2 or R2* depending on data
acquisition technique
• Characteristic time of decay known as T2 or T2*
depending on data acquisition technique
Clark PR, St. Pierre TG. Mag Res Imaging. 2000;18:431-438.
Calculating Tissue Iron from MRI
Measurements
Relaxometry methods, eg R2 or R2*
Intensity ratio methods
The rate at which
signal decays is known
as R2 or R2*
100
Signal Strength
Typical non–iron-loaded tissue
The characteristic
time of decay is
known as T2 or T2*
80
60
Effect of
increasing
iron loading
40
20
0
0
5
10
15
20
Echo Time (ms)
St. Pierre TG. Ann N Y Acad Sci. 2005;1054:379-385. Graphic courtesy of Dr. Tim St. Pierre.
Methods of Measurement of Tissue
Iron Concentrations with MRI
• Relaxometry measurement of R2 is the most
widespread method for measurement of liver
iron concentration
• Relaxometry measurement of T2* is the most
widespread method for assessing iron in the
heart
Proton Transverse Relaxation Rate
(R2) Image and Distribution
LIC = 7.7 mg.g-1
R2 (s-1)
Transverse Relaxation Rate R2 (s-1)
With permission from St. Pierre TG, et al. Blood. 2005;105:855-861.
Liver R2 Images and Distributions
• Non–iron-loaded subject
• 3 iron-loaded subjects
• R2 distribution shifts to
higher values as LIC
increases
With permission from St. Pierre TG, et al. Blood. 2005;105:855-861.
Mean Transverse Relaxation Rate <R2> (s-1)
Dissected Liver Samples
Mean R2 vs iron
concentration for 32
cubes of liver dissected
from a single iron–loaded
liver postmortem
Iron Concentration (mg/g dw)
With permission from Clark PR, et al. Mag Res Med. 2003;49:572–575.
Mean Transverse Relaxation Rate <R2> (s-1)
Relationship Between R2 and
Needle Biopsy LIC (dw)
Biopsy Iron Concentration (mg/g dry tissue)
With permission from St. Pierre TG, et al. Blood. 2005;105:855-861.
R2-MRI Is a Reliable Measure of LIC
• High sensitivity and
specificity over entire range
of LIC encountered
• Unaffected by presence of
fibrosis/cirrhosis
Fibrosis stages:
0–1 = 
2–4 = 
5–6 = 
With permission from St. Pierre TG, et al. Blood. 2005;105:855-861.
Example—R2-MRI Measurements to
Monitor Iron Chelation Therapy
Before chelation therapy
intervention
Mean LIC = 16.0
After 12 months of
chelation therapy
intervention
Mean LIC = 1.6
LIC map
Low
iron
Graphic courtesy of Dr. Tim St. Pierre.
High
iron
Methods of Monitoring Heart Iron
Methods for Heart Iron Assessment
T2* methods are used to assess heart iron loading
Echo time increasing
With permission from Westwood M, et al. J Magn Reson Imaging. 2003;18:33-39.
Relationship Between T2* and
Cardiac Function
With permission from Anderson LJ, et al. Eur Heart J. 2001;22:2171-2179.
Relationship Between R2* and
Cardiac Function
Transform using R2* = 1/T2*
100
With permission from Anderson, LJ, et al.
Eur Heart J. 2001;22:2171-2179.
LVEF (%)
80
60
40
20
0
0
50
100
150
200
250
-1
R2* (s )
Abbreviation: LVEF, left ventricular ejection fraction.
Graphic courtesy of Dr. Tim St. Pierre
Calibration of Cardiac T2*/R2*
Against Tissue Iron Concentration
Preliminary calibration
over small iron
concentration range
obtained from a single
human heart
With permission from Ghugre, et al. Magn Reson Med. 2006;56:681-686.
Implementing These Methods at
Your Institution
Implementing These Methods at
Your Institution
• MRI data acquisition
– Relatively simple for liver
– More involved for heart
 Requires extra hardware and software on scanner
• MRI data analysis
– Problematic for liver
 High risk of erroneous analysis due to low signal to noise
ratios; need to account for background noise, etc.
– Relatively simple for heart
Implementing These Methods at
Your Institution
• MRI data acquisition
– Liver
 No face-to-face training required
– Heart
 May require expert training of technicians
• MRI data analysis
– Liver
 ISO9001 Quality Assurance should be implemented, or data
analysis should be outsourced to quality assured core lab
– Heart
 Technicians should receive training from experts
When to Measure Iron in the Liver
vs the Heart
• Patients on regular blood transfusion
– Measure liver iron annually
– Measure heart iron annually after 20 units have been
transfused
• Patients with hereditary haemochromatosis
– Measure liver iron at diagnosis if >40 years of age and
serum ferritin >1000 ng/mL
• Patients with thalassaemia intermedia
– Measure liver and heart iron annually after age 10 years
– If the baseline cardiac T2* in normal range, subsequent
cardiac T2* no more frequent than 3–5 years unless
there is difficulty controlling the liver iron
Conclusions
• It is now possible in most major hospitals
to monitor iron in the liver and the heart
using magnetic resonance imaging
• The ability to measure iron in these 2
organs provides the basis for making
better informed decisions concerning the
need to adjust patients’ chelation regimens