British Journal of Radiology 75 (2002),24-27 © 2002 The British Institute of Radiology
The effect of haemosiderosis and blood transfusions on the T2 relaxation time and 1/T2 relaxation rate of liver tissue
S Salo, MD1,
A Alanen, MD1,
R Leino, MD2,
S Bondestam, MD3 and
M Komu, PhD1
Departments of 1Diagnostic Radiology and 2Medicine, Turku University, FIN 20520 Turku and 3Department of Diagnostic Radiology, Helsinki University Central Hospital, FIN 00290 Helsinki, Finland
Correspondence: Sakari Salo, MD, Department of Diagnostic Radiology, Turku University Hospital, PL 52, FIN-20521 Turku, Finland
 |
Abstract
|
|---|
Patients with chronic anaemia need repeated blood transfusions, which eventually lead to iron overload. The excess iron from blood transfusions is deposited in the reticuloendothelial system and in the parenchymal cells of the liver, spleen and other organs. Cellular damage is likely to occur when iron overload in the liver is pronounced. Liver biopsy is still necessary to evaluate the degree of haemosiderosis or haemochromatosis. To avoid this invasive procedure, methods have been sought to determine the concentration of iron in liver tissue and to estimate the effect of the treatment of haemosiderosis or haemochromatosis. In this MRI study, the T2 relaxation time and the 1/T2 relaxation rate of liver were determined in 23 patients who had undergone repeated blood transfusions for chronic anaemia. The first 60 transfusions had the greatest influence on the measured T2 relaxation time, with T2 relaxation time decreasing as haemosiderosis progresses. The 1/T2 relaxation rate increases significantly in a linear fashion when the number of blood transfusions increases up to 60. After 60 transfusions the influence of additional blood transfusions on the T2 value was minimal; the same response, although in reverse, was seen in the 1/T2 relaxation rate curve. One possible explanation for this may be that the MR system could detect the effect of only a limited amount of iron excess and any concentration over this limit gives a very short T2 relaxation time and a very weak signal from the liver, which is overwhelmed by background noise. However, in mild and moderate haemosiderosis caused by blood transfusions, T2 relaxation time and 1/T2 relaxation rate reflect iron accumulation in liver tissue.
|
Introduction
|
|---|
Patients with chronic anaemia require repeated blood transfusions as part of the treatment of their underlying disease. The excess iron from blood transfusions tends to be deposited in the reticuloendothelial system of the liver, spleen and bone marrow. The parenchymal cells of the liver also store surplus iron [1]. Repeated blood transfusions eventually lead to transfusional haemosiderosis of the liver, while severe iron overload may cause hepatic fibrosis and cirrhosis [1–3].
The iron concentration in the liver can be determined from histological samples. However, the biopsy procedure is invasive and predisposes to complications. Alternative methods of determining the iron concentration in the liver have therefore been sought. Parameters such as serum ferritin, serum iron and transferrin saturation are widely used but are indirect methods that evaluate total body iron stores and not liver iron stores. While serum ferritin correlates well with liver iron concentration in healthy subjects, this correlation is lost in some diseases [4–8]. A similarly poor correlation between serum ferritin and liver iron concentration is found with hereditary haemochromatosis during venesection therapy [9] and with chronic anaemia treated by multiple blood transfusions [10]. Liver iron concentration can be detected as an increase in liver attenuation using CT, but the sensitivity of this method is poor despite its good specificity [11, 12]. Several studies have shown that MRI reliably shows iron overload of the liver [13–19]. The accumulation of iron in the liver changes the T2 relaxation time of the liver, which is seen on both T1 and T2 weighted MR images as a decrease in signal intensity [14, 19, 20]. Furthermore, the 1/T2 relaxation rate shows a nearly linear correlation with liver iron concentration [16].
While it has been shown that liver iron concentration correlates better with 1/T2 relaxation rate than with signal intensity [17], the opposite results have also been published [15]. However, different sequences were used in these studies, which may explain the difference in results.
By analysing histological samples, Risdon et al [1] found a correlation between liver iron concentration and the total number of units of blood transfused. Gomori et al [21] found that it was possible to estimate the quantity of iron in liver tissue by measuring the 1/T2 relaxation rate.
The purpose of the present study was to determine whether the number of blood transfusions, i.e. when the incoming quantity of iron is possible to estimate, correlates with the measured T2 relaxation time and 1/T2 relaxation rate of the liver tissue.
|
Methods and materials/patients
|
|---|
MRI was performed with a high field 1.5 T MRI system (Magnetom 63SP; Siemens, Erlangen, Germany) using standard software and a body coil. Using the software package of the imager, the liver T2 relaxation times were calculated from the data obtained from a specific Carr–Purcell–Meiboom–Gill pulse sequence with TR 2000 ms and TE 20 ms+n x 15 ms (where n is in the range 0–15). The slice thickness was 10 mm, matrix size 128 x 256, field of view 500 mm and acquisition time 8 min 37 s.
The liver T2 values each represent the mean of four operator-defined regions of interest with an area of 430–770 mm2. Measured areas were selected to avoid visible vessels and regions with artefact.
Normal liver T2 relaxation time and 1/T2 relaxation rate were determined from the livers of 19 volunteers (9 men and 10 women aged 18–75 years) to ascertain whether T2 relaxation time or 1/T2 relaxation rate change with age. There was no previous history of blood transfusions or liver-related diseases among the volunteers over their entire lifetime. Five volunteers used anti-inflammatory drugs. There was also one volunteer who consumed 7–14 g of alcohol per day.
MRI was performed on 23 patients who had undergone repeated blood transfusions for various reasons (9 men and 14 women aged 20–84 years, mean age 55 years; Table 1
), to determine the T2 relaxation time and 1/T2 relaxation rate of the liver.
View this table:
[in this window]
[in a new window]
|
Table 1. Total number of blood transfusions (NBT), T2 relaxation time of liver tissue and diagnosis of the patients
|
|
The linear regression method was applied to calculate the behaviour of 1/T2 of liver as a function of the number of blood transfusions (NBT).
The T2 relaxation time of the spleen (11 volunteers, 22 patients), subcutaneous fat (19 volunteers, 26 patients) and arm muscle (biceps ortriceps) (16 volunteers, 16 patients) was also calculated when possible. Measurements were done from one operator-defined region in each tissue, the area of which varied from 80–310 mm2.
|
Results
|
|---|
The age-related variation of normal liver T2 relaxation time is presented in Figure 1. The calculated T2 relaxation time±standard deviation (SD) of the mean of normal human liver tissue was 51.7±5.0 ms (95% confidence interval (CI) 49.4–53.9 ms). Age had no discernable influence on T2 relaxation time, with no correlation (r=-0.0898) between age and T2 relaxation time of the liver tissue. In our study, the range of mean T2 values for normal liver tissue was 44.8–59.9 ms. The relaxation rate 1/T2 of normal human liver tissue was 19.5±1.9 s-1 (95% CI 18.7–20.4 s-1).
View larger version (9K):
[in this window]
[in a new window]
|
Figure 1. The T2 relaxation time of the liver does not change significantly with age. Correlation between T2 relaxation time and age is -0.090.
|
|
T2 and 1/T2 values of the liver against NBT are presented in Figures 2 and 3, respectively. The measured relaxation rate 1/T2 of the liver tissue increased linearly (slope=1.31±0.30 1/sNBT, r=0.737, p=0.0005) when the NBT increased up to 60 (Figure 3). After that, the influence of additional blood transfusions on the measured T2 value was minimal and the linear correlation vanished. The value of 1/T2 is 11.8±7.1 s-1 when the NBT is zero.
View larger version (12K):
[in this window]
[in a new window]
|
Figure 3. The 1/T2 relaxation rate increases when the number of blood transfusions (NBT) increases. The best correlation between 1/T2 and NBT was found when the NBT was less than 60. Linear regression (—) with 95% confidence interval (- - -) is represented when NBT <60.
|
|
Calculated T2 relaxation times±SD (95% CI) of the normal spleen, subcutaneous fat and normal arm muscle (biceps or triceps) were 92.4±11.3 ms (86.1–98.3 ms), 115.2±7.2 ms (112.3–118.6 ms) and 38.9±7.4 ms (35.4–42.3 ms), respectively. Values for patients were 35.3±13.5 ms (29.7–41.0 ms), 115.1±4.8 ms (113.3–117.0 ms) and 39.6±5.3 ms (37.0–42.2 ms), respectively. The T2 relaxation time of spleen changed significantly, but there was no correlation between the NBT and T2 relaxation time (r=0.0999). The range of T2 relaxation times of patients' spleens was 10.3–57.1 ms. T2 relaxation times of fat and muscle did not differ between patients and volunteers.
|
Discussion
|
|---|
A slight variation in the iron distribution of iron overloaded liver tissue has been detected in earlier studies [21, 22]. Therefore T2 relaxation time was measured from four equally sized areas of the liver to avoid misinterpretation caused by this phenomenon.
The T2 relaxation time of normal liver tissue in the control group remained the same despite differences in age. Therefore, aging could not account for any significant change in the T2 relaxation time or 1/T2 relaxation rate of the liver.
In the repeated blood transfusions group, blood transfusions clearly shortened the T2 relaxation time and increased the 1/T2 relaxation rate of liver tissue. Earlier studies have shown that a decrease in T2 relaxation time or in signal intensity reflects iron accumulation in liver tissue, with good correlation [19, 20]. In this study, the measured 1/T2 relaxation rate of the liver of patients increased in a linear manner during the first 60 blood transfusions. The influence of additional transfusions on the T2 value was minimal. Other authors have observed a corresponding decrease in signal intensity of the liver on both T1 and T2 weighted images when the iron concentration has increased in the liver tissue [14, 15]. One explanation for this minimal change with abundant blood transfusions may be that the MR system can detect only a limited amount of iron excess. Concentrations over this limit cause very short T2 values and noise overwhelms the measured signal from the liver on T2 weighted images. Moreover, Risdon et al [1] showed that liver iron concentration showed a good correlation with the amount of blood received until about 100 units (not 60 units) of blood had been given, while thereafter the observed iron concentrations were progressively less than expected for the transfused load. Their study was based on histological samples and strengthens the suspicion of the capability of MRI to detect differences in tissue iron concentration with severe iron overload. The T2 relaxation time of heavily transfused liver was considerably shorter than the minimum TE of 20 ms used in our study, which could explain at least in part the differences between these two studies. Further examinations are needed to clarify this aspect. However, it seems that in some cases, measured 1/T2 or T2 values could be useful when considering the need for chelation therapy.
One interesting finding was that T2 relaxation time of the spleen changed significantly after blood transfusions, but the level of T2 relaxation time remained almost the same despite the change in NBT. It therefore seems possible that the iron saturation point of the spleen is already reached during the first blood transfusions.
In conclusion, our study shows that MR examination reflects iron accumulation in liver tissue in mild and moderate haemosiderosis caused by blood transfusions. At that stage, possible tissue damage (fibrosis and cirrhosis) is still insignificant [1] and further development of tissue damage might be preventable.
|
Acknowledgments
|
|---|
This study has been supported by Medical Research Foundation of Turku University Central Hospital and the Finnish Foundation for Alcohol Studies.
Received for publication January 18, 2001.
Revision received August 29, 2001.
Accepted for publication September 4, 2001.
|
References
|
|---|
-
Risdon RA, Barry M, Flynn DM. Transfusional iron overload: the relationship between tissue iron concentration and hepatic fibrosis in thalassaemia. J Pathol 1975;116:83–95.[Medline]
-
Powell LW, Bassett ML, Halliday JW. Hemochromatosis: 1980 update. Gastroenterology 1980;78:374–81.[Medline]
-
Schafer AI, Cheron RG, Dluhy R, Cooper B, Gleason RE, Soeldner JS, et al. Clinical consequences of acquired transfusional iron overload in adults. N Engl J Med 1981;304:319–24.[Abstract]
-
Bentley DP, Williams P. Serum ferritin concentration as an index of storage iron in rheumatoid arthritis. J Clin Pathol 1974;27:786–8.[Abstract/Free Full Text]
-
Birgegård G, Hällgren R, Killander A, Strömberg A, Venge P, Wide L. Serum ferritin during infection. A longitudinal study. Scand J Haematol 1978;21:333–40.[Medline]
-
Brissot P, Bourel M, Herry D, Verger JP, Messner M, Beaumont C, et al. Assessment of liver iron content in 271 patients: a reevaluation of direct and indirect methods. Gastroenterology 1981;80:557–65.[Medline]
-
Jones PA, Miller FM, Worwood M, Jacobs A. Ferritinaemia in leukaemia and Hodgkin's disease. Br J Cancer 1973;27:212–7.[Medline]
-
Prieto J, Barry M, Sherlock S. Serum ferritin in patients with iron overload and with acute and chronic liver diseases. Gastroenterology 1975;68:525–33.[Medline]
-
Jensen PD, Jensen FT, Christensen T, Ellegaard J. Non-invasive assessment of tissue iron overload in the liver by magnetic resonance imaging. Br J Haematol 1994;87:171–84.[Medline]
-
Johnston DL, Rice L, Vick GW, Hedrick TD, Rokey R. Assessment of tissue iron overload by nuclear magnetic resonance imaging. Am J Med 1989;87:40–7.[Medline]
-
Chezmar JL, Nelson RC, Malko JA, Bernardino ME. Hepatic iron overload: diagnosis and quantification by noninvasive imaging. Gastrointest Radiol 1990;15:27–31.[Medline]
-
Guyader D, Gandon Y, Deugnier Y, Jouanolle H, Loreal O, Simon M, et al. Evaluation of computed tomography in the assessment of liver iron overload. A study of 46 cases of idiopathic hemochromatosis. Gastroenterology 1989;97:737–43.
-
Bassett ML, Halliday JW, Powell LW. Value of hepatic iron measurements in early hemochromatosis and determination of the critical iron level associated with fibrosis. Hepatology 1986;6:24–9.[Medline]
-
Bondestam S, Lamminen A, Anttila VJ, Ruutu T, Ruutu P. Magnetic resonance imaging of transfusional hepatic iron overload. Br J Radiol 1994;67:339–41.[Abstract]
-
Gandon Y, Guyader D, Heautot JF, Reda MI, Yaouanq J, Buhé T, et al. Hemochromatosis: diagnosis and quantification of liver iron with gradient-echo MR imaging. Radiology 1994;193:533–8.[Abstract/Free Full Text]
-
Kaltwasser JP, Gottschalk R, Schalk KP, Hartl W. Non-invasive quantitation of liver iron-overload by magnetic resonance imaging. Br J Haematol 1990;74:360–3.[Medline]
-
Papakonstantinou OG, Maris TG, Kostaridou V, Gouliamos AD, Koutoulas GK, Kalovidouris AE, et al. Assessment of liver iron overload by T2-quantitative magnetic resonance imaging: correlation of T2-QMRI measurements with serum ferritin concentration and histologic grading of siderosis. Magn Reson Imaging 1995;13:967–77.[Medline]
-
Rocchi E, Cassanelli M, Borghi A, Paolillo F, Pradelli M, Casalgrandi G, et al. Magnetic resonance imaging and different levels of iron overload in chronic liver disease. Hepatology 1993;17:997–1002.[Medline]
-
Stark DD, Moseley ME, Bacon BR, Moss AA, Goldberg HI, Bass NM, et al. Magnetic resonance imaging and spectroscopy of hepatic iron overload. Radiology 1985;154:137–42.[Abstract/Free Full Text]
-
Stark DD, Bass NM, Moss AA, Bacon BR, McKerrow JH, Cann CE, et al. Nuclear magnetic resonance imaging of experimentally induced liver disease. Radiology 1983;148:743–51.[Abstract/Free Full Text]
-
Gomori JM, Horev G, Tamary H, Zandback J, Kornreich L, Zaizov R, et al. Hepatic iron overload: quantitative MR imaging [see comments]. Radiology 1991;179:367–9.[Abstract/Free Full Text]
-
Kadoya M, Matsui O, Kitagawa K, Kawamori Y, Yoshikawa J, Gabata T, et al. Segmental iron deposition in the liver due to decreased intrahepatic portal perfusion: findings at MR imaging. Radiology 1994;193:671–6. [Erratum. Radiology 1995;194:915.][Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
G. M. Brittenham and D. G. Badman
Noninvasive measurement of iron: report of an NIDDK workshop
Blood,
January 1, 2003;
101(1):
15 - 19.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
