Thalassaemia International Federation | A non-profit, non-governmental patient-driven organization, founded in 1986

Magnetic Resonance Imaging (MRI) is a key tool in monitoring thalassaemia patients, particularly in order to assess iron loading in major organs and to provide early warning of when iron chelation regimes need to be introduced or modified.

The images produced by 1.5 Tesla magnets allow estimation of iron load in the liver and heart, providing useful endpoints in monitoring the various regimes that the three chelating agents available to clinical practice now offer.

The liver was the first organ to be investigated by MRI, and this has been adopted as a routine method in many centres, aimed at replacing liver biopsy for assessing liver iron concentration (LIC). Various methods of quantification investigated over the previous decade are now being used in routine clinical practice.

T2* imaging of the heart reveals the degree of iron loading and has revolutionised the management of thalassaemia. The technique enables medical staff to detect cardiac deposition early and to adjust chelation therapy accordingly, thereby improving iron load and – importantly – heart function, so having an effect on cardiac mortality.

Another organ in which MRI has been investigated is the pituitary gland, one of the earliest “targets” of iron overload. However, its utility in clinical practice has not yet been established.

The latest organ to be investigated is the pancreas. This organ is also a target for iron deposition with a danger for -cell damage and the development of diabetes. Indeed, diabetes is a well recognised complication, especially in adult patients with thalassaemia.

In a multi-centre study organised by TIF2 to investigate the prevalence of endocrine complications in 29 centres treating 3,817 patients with -thalassaemia across the globe, impaired glucose tolerance was found in 6.5% and insulin-dependent diabetes mellitus was found in 3.2%.

MRI imaging is already proving useful in type I diabetes, unrelated to iron overload. In non-thalassaemia patients with type I diabetes, there is alteration in vascular flow, volume and permeability. These alterations in micro-vascular dynamics can be investigated using specialised contrast agents making MRI imaging useful in the early stages of the disease.3 In the course of the disease pancreatic volume, which shows a reduction of almost 50% in long-standing diabetes, can be monitored by MRI.

In thalassaemia, these measurements of vascular flow have not yet been assessed, but pancreatic iron deposition and pancreatic volume are the subjects of investigation. The first technique to be investigated was ultra-sonography,5 which shows increased echogenicity and decreased size of the gland due to iron deposition. This finding is associated with a disturbance of function and so may be regarded as an early index of an increased risk for developing diabetes.

In the earlier studies on MRI imaging of the pancreas7 the major finding was that iron deposition, indicated by a hypo-intense signal, progressed to a hyper-intense image, indicating fatty replacement of the parenchyma correlated with impairment of exocrine pancreatic function, and that this fatty replacement may be associated with glucose disturbances and so be predictive of diabetes. Patients with pathologic oral glucose tests (OGTT) had higher liver siderosis, but also fatty replacement of the liver.

Recent work9 has also confirmed the possible usefulness of MRI in predicting diabetes. The investigation used T2* signals to correlate with pre-diabetic biochemical findings. Some important observations were recorded in this study:

Haemosiderosis of the pancreas is not related to age or ferritin levels. Pancreatic haemosiderosis does not correlate with liver haemosiderosis, but correlates strongly with cardiac siderosis. Chelating agents such as desferrioxamine, which may fail to prevent cardiac siderosis, may also fail to protect the pancreas. Also, the results of cardiac T2* may be a “surrogate” assessment of pancreatic siderosis, so that routine clinical practice that includes cardiac MR imaging may not require pancreatic imaging. The pancreatic volume, if reduced, indicating atrophy, is a better predictor of diabetic status than pancreatic haemosiderosis. This confirms the findings in non-thalassaemic patients with type I diabetes4 but may be an indicator of long-standing diabetes rather than early, reversible condition.

From these studies it would appear that the onset of disturbances of glucose metabolism, a pre-diabetic state, may be predicted by MRI imaging of the pancreas if there is:
a)fatty replacement of the pancreatic parenchyma;
b)pancreatic siderosis using T2*;
c)cardiac siderosis using T2*; and
d)pancreatic volume changes (atrophy).

The importance of these findings is that MRI may be a useful tool in predicting diabetes, before established disease appears. This is significant only if impaired glucose metabolism can be reversed by intensifying chelation therapy. There is evidence that this may be the case. Most recent publications have used combination therapy (desferrioxamine and deferiprone) and positive results in a proportion of patients have been reported.10,11,12 Established insulin-dependent diabetes is irreversible, so that the predictive tests are invaluable if they can avoid further complications in the lives of patients with thalassaemia. However, more studies are needed – to assess MRI techniques and their value, as well as the effects of intensive iron chelation in this complication.

References:

1) Tanner MA, Galanello R, Dessi C, Smith GC, Westeood MA, et al. A randomized, placebo controlled, double-blind trial of the effect of combined therapy with Desferrioxamine and Deferiprone on myocardial iron in thalassaemia major using cardiovascular magnetic resonance. Circulation, 2007, 115(14): 1876-84.
2) De Sanctis V, Eleftheriou A, Malaventura C. Prevalence of endocrine complications and short stature in patients with thalassaemia major: A multi-centre study by the Thalassaemia International Federation (TIF). Paediatr. Endocrinol. Rev., 2004, Suppl. 2: 249-55.
3) Medorova Z, Castillo G, Dai G, Bolotin E, Bogdanov A, Moore A. Non-invasive magnetic resonance imaging of microvascular changes in type I diabetes. Diabetes, 2007, 56(ii): 2677-82.
4) Williams AJ, Chan W, Callaway MP, Dayan CM. Magnetic Resonance Imaging: A reliable method for measuring pancreatic volume in type I diabetes. Diabet. Med., 2007, 24(1): 35-40.
5) Gullo L, Corcioni E, Brancati C, Brian M, Peizzili R et al. Morphologic and functional evaluation of the exocrine pancreas in beta-thalassaemia major. Pancreas, 1993, 8(2): 176-80.
6) Theochari M, Ioannidou D, Nounopoulos H, Bouloukos A, et al. Ultrasonography of the pancreas as a function index in children with beta thalassaemia. J Pediatr Endocrinol Metab, 2000, 13(3): 303-6.
7) Midiri M, Lo Casto A, Sparacia G, D’Angelo P et al. MR Imaging of pancreatic changes in patients with transfusion dependent β–thalassaemia major. Amer J Radiol, 1999, 173:187-192.
8) Papakonstandinou U, Ladis V, Kostaridou S, Maris T, Berdousi H, Kattamis C, Gourtsoyiannis N. The pancreas in beta-thalassaemia major: MR Imaging features and correlation with iron stores and glucose disturbances. Eur Radiol, 2007, 17(6): 1535-43.
9) Au WY, Lam WW, Chu W, Tam S, Wong WK, Liang R, Ha SY. A T2* magnetic imaging study of pancreatic iron overload in thalassaemia major. Haematologica, 2008, 93(1): 116-9.
10) Platis O, Anagnostopoulos G, Farmaki K, Posantzis M, Gotsis E, Tolis G. Glucose metabolism disorders improvement in patients with thalassaemia major after 24-36 months of intensive chelation therapy. Pediatr Endocrinol Rev, 2004, suppl 2:270-81.
11) Farmaki K, Angelopoulos N, Anagnostopoulos G, Gotsis E, Rombopoulos G, Tolis G. Effect of enhanced iron chelation on glucose metabolism in patients with beta-thalassaemia. Br J Haematol, 2006, 134(4): 438-44.
12) Christoforidis A, Perifanis V, Tsarta I, Vlachaki E, Athanassiou-Metaxa M. Evolution of OGTT in patients with beta thalassaemia major in relation to chelation therapy. Diabetes Res Clin Pract, 2007, 76(1): 6-11.

Medical focus

From these studies it would appear that the onset of disturbances of glucose metabolism, a pre-diabetic state, may be predicted by MRI imaging of the pancreas if there is:

References: