British Journal of Haernatology, 1995.91, 827-833

Liver iron stores in patients with secondary haemosiderosis under iron chelation therapy with deferoxamine or deferiprone

P. NIELSEN, R . FISCHER, R. ENGELHARDT, P. TONDURY,* E. E. GABBE A N D G . E. J A N K A ~ Abt. Medizinische Biochemie, Physiologisch-Chemisches lnstitut, tHiimatol. und Onkol. Abteilung, Kinderklinik Universitiitskrankenhaus Eppendorf, Hamburg, Germany, and *Kinderklinik, Znselspital Bern, Schweiz

Received 24 February 1995; accepted for publication 3 August 1995

Summary. Total body iron stores including liver and spleen iron were assessed by nod-invasive SQUID biomagnet- ometry. The liver iron concentration was measured in groups of patients with ,O-thalassaemia major or other post- transfusional siderosis under treatment with the oral iron chelator deferiprone (n = 19) and/or with parenteral defer- oxamine (n = 33). An interquartile range for liver iron concentrations of 1680-4470pg/g liver was found in these patients. In both groups a poor correlation between liver iron and serum ferritin values was observed.

Repeated measurements of liver and spleen iron concen- trations as well as determination of liver and spleen volume by sonography were performed in six patients under continuous deferiprone treatment for 3-15 months. In this group detailed information was obtained on the whole body iron store (5-36 g) and the iron excretion rates (14-34 mg/d)

for each patient. As indicated by decreasing liver iron concentrations, five out of six subjects showed a negative iron balance (2 - 1 3 mg/d). Conventional measurements of both serum ferritin and urine iron excretion gave fluctuating results, thus being only of limited use in the control of iron depletion therapy.

The non-invasive biomagnetic liver iron quantification is a precise and clinically verified technique which offers more direct information on the long-term efficacy of an iron depletion therapy than the hitherto used methods. This technique may be of use in the clinical evaluation of new oral iron chelators.

Keywords: thalassaemia. biomagnetometry, iron, L1, deferiprone.

In patients with secondary iron overload (e.g. thalassaemia major), the degree of iron loading is a factor which is correlated with the expectation of life. A severe siderosis can result in cardiomyopathy, diabetes or liver cirrhosis. The iron chelator deferoxamine (DFO) has been shown to deplete increased iron stores and prolong life (Pippard et al, 1982). Preliminary data indicate that the oral iron chelator deferiprone (DFP, former names L1, CP20) may have similar effectiveness in promoting iron excretion in iron-loaded patients (Kontoghiorghes et al, 1987a, 1 9 9 0 Olivieri et al, 1990). The efficacy of iron chelation therapy is usually judged by measuring urine iron excretion or changes in parameters of iron metabolism, e.g. serum ferritin. However, these parameters give no reliable information on the amount and changes of the storage iron under treatment especially

Correspondence: Dr Peter Nielsen, Abt. Medizinische Biochemie. Institut fiir Physiologische Chemie, Universitatskrankenhaus Eppen- dorf, Martinistr. 52,20246 Hamburg, Germany.

0 1995 Blackwell Science Ltd

in severely iron-loaded patients (Worwood et al, 1980; Brittenham et al, 1993).

Some non-invasive techniques, such as biomagnetic liver susceptometry (BLS) (Brittenham et al. 1982, 1993; Fischer et al, 1989, 1992) or quantitative magnetic resonance imaging (Kaltwasser et al, 1990; Engelhardt et al, 1994), are able to measure liver iron concentrations in iron-loaded patients precisely and may offer a new way to monitor the therapeutic efficacy of an iron-depletion therapy more directly. In the following report we used BLS to monitor iron chelation therapy by DFO and DFP in 52 patients with severe secondary iron overload. We report the results of a prospective study about the efficacy of deferiprone in six of these patients. Moreover, two of these patients with ,O- thalassaemia major could be monitored for the first time by biomagnetometry from the beginning of their chelation therapy with alternating doses of DFO and DFP, to nearly equilibrium liver iron concentrations after more than 3 years.

82 7

828 P. Nielsen et a1 Table I. Biochemical parameters in six patients with severe iron overload before therapy with deferiprone.

Pt

1 2 3 4 5 6

- Diagnosis

M D S ~ O M F ~ P-thal.

P-thal. P-thal.

P-thd.

Liver Fe (mg/g liver)

4.0 f 0.2 5.3 f 0.5 9.8 f 0.2 5.1 f 0.3 5.2 f 0.2 4.4 f 0.2

Liver volume (ml)

3414 2150 3280 950 920

1050

Spleen Fe (ms/g spleen)

2.2 f 0.2 - - 2.9 f 0.2 2.0 f 0.3

(2-215

Spleen volume (ml)

2850 - - 320 210 160

Total iron store3 (9)

21.9 f 1.5 12.7 f 1.4 35.7 f 1.2 9.7 f 0.6 9.1 f 1.0 5.5 & 0.3

Serum ferritin (Pdl)

6978 5400

12900 3230 2590 2900

Myelodysplastic syndrome: * osteomyelofibrosis; calculated acc. to eq. (1); ALT = alanine aminotransferase; 50% of liver Fe assumed.

PATIENTS AND METHODS

1,2-Dimethyl-3-hydroxypyrid-4-one (DFP) was synthesized as described (Hider et aI, 1982: Kontoghiorghes 81 Sheppard, 198 7b). After 5-fold recrystallization from water, the purity of each batch was confirmed by HPLC (column, PrepRPC, Pharmacia; eluant, l O m ~ NaH2P04, l m ~ EDTA, 2% acetonitrile.

Patients with 0-thalassaemia major (n = 34), thalassae- mia intermedia (n = 5) or with other post-transfusional haemosiderosis being in out-patient medical attendance in the University Hospital Hamburg, Germany, or Inselspital Bern, Switzerland, were sent to the iron metabolism centre of the University Hospital Hamburg for non-invasive liver iron quantification by SQUID biomagnetometry.

All patients under DFP treatment gave their free and informed written consent to be treated with this compound. Several patients were measured twice, during their DFO therapy and 6-12 months later before starting DFP therapy. A group of six patients (see Table I) were studied in time intervals of 2-6 months during their DFP therapy lasting from 3 to 15 months. Due to severe local reactions against DFO infusions two of these six patients were receiving alternating treatment with DFO (1.7 g/d) and DFP (1.4 g/d) for 6 months each. Biomagnetic susceptibility measurements

of liver and spleen were performed within 3 weeks of drug change. Appropriate doses (see Table 11) were administered in gelatine capsules, >30min before a meal. Transfused erythrocyte concentrates contained on average 243 ml, corresponding to 24.5 g Hb/dl= 0,849 mg Fe/ml. Iron in urine was measured by atomic absorption spectroscopy (AAS) with graphite furnace (Perkin Elmer 2100). In order to correct for incomplete collection during 24-48 h periods, daily urinary iron excretion was normalized to standard potassium excretion instead of creatinine excretion. Potas- sium was measured by germanium spectroscopy via its isotopic 40K y-radiation.

Estimation of total body iron stores using biornugnetornetry. Liver and spleen iron concentrations were measured using a SQUID biomagnetometer (Ferritometer: BTi, U.S.A.) as described elsewhere (Fischer et aJ, 1989: Paulson et al, 1991). Patients in supine position with a water coupling membrane above the liver or spleen are lowered in the magnetic field of a superconducting coil (20mT). The magnetic flux change (2 100 pT) during the movement is registered by two detector coils (R = 3, 7 cm) for 1 3 s. SQUID output voltages are analysed as a linear function of the individual magnetization geometry resulting in magnetic volume susceptibilities, Ax = x - xwater. for the overlying body tissue and the liver (spleen). The liver iron

Table 11. Efficacy of iron depletion under treatment with deferiprone: mean daily total body elimination rates (mg Fe/d) as measured by biomagnetic liver susceptometry.

DFP DFP (DFO) Serum ferritin Urine treatment dose trend' Fe excretion Norm. to 40K F b f d , , Febalance * Feexcretim '

Pt (4 (mg/kg/d) (pg/l/d) (mg/d) (mg/d) (mg/d) (mdd) (mg/d)

1 196 75 +11 f 5 7 k 3 6 f 2 54.7 $20.4 f 2.5 34.3 2 44 1 84 -2 f 1 1 6 f 6 2 1 f 7 19.6 -12.8 & 1.6 32.4 3 393 77 +2 f 6 8 3 f 1 8 5 7 f 9 18.9 -5.3 f 1.6 24.2 4 1 804 74 (90) -11 f 9 4 k 2 8 4 ~ 3 12.7 -6.4 f 0.6 19.1

6 85 70 -10 f 25 6 f l l l f l 12.2 -2.2 f 0-3 14.4 5 1804 70 (85) -14 f 5 4 & 2 10f 1 11.7 - 5 ' 7 f 1.0 17.4

Linear regression of serum ferritin values under DFP treatment: measured changes in total body iron stores (liver + spleen): ' calculated from Fetrmsfusion - F%alance; mean duration of DFP/DFO therapy intervals (see Fig 3) .

0 1995 Blackwell Science Ltd, British Journal ofHaematoZogy 91: 827-833

Biomagnetometry in Iron Chelation Therapy 829 iron excretion rate can be derived from equation ( 2 ) by the differential dUex/dt = k2 - Us(t). For larger therapy periods, the model independent integral over t - to (eq. (3)) has to be compared with urinary iron excretion:

Ue.,(t - t o ) = K1 ( t - to) - [Us(t) - Uol (3) An iron balance rate, F%dance = [Us(t) - Uo]/(t - to) C 0,

will indicate an effective chelation therapy with respect to a certain transfusion rate.

DFT DFO

1-1 I / BodyIron Store I

Fig 1. Two-compartment model for changes of body iron stores under blood transfusions and iron chelator treatment: K1 = rate of transfused iron: k2 = total body iron excretion rate constant.

concentration is calculated as cp, = Axliver/( with the specific volume susceptibility, < = 1600 x m3/kgFe for the paramagnetic ferritin/haemosiderin complex (Shoden & Sturgeon, 1960). The spleen iron concentration is deter- mined in the same way, usually in an elliptic magnetization geometry. The individual liver and spleen geometry is assessed by sono-graphy. Thus, main total body iron stores can be estimated by measuring liver and spleen volumes, V, by bedside sonographic imaging in a similar technique as described by Leung et a1 (1986). Laser beam positioned sagittal slices at known gap distances are integrated by the imager software (CS9500, Picker, Germany). Assuming 90% of storage iron in liver and spleen for iron overload patients, the total body iron store, Us, is calculated from biomagne- tically determined iron concentrations, cFe, of liver and spleen according to equation (1).

Us = [ ( c F ~ * O i v e r + ( C F ~ * V)sp~J/O-9 (1)

The body iron stores during a certain therapy time interval, t > to, are interpreted in the framework of a two- compartment model as outlined in Fig 1. The transfused iron, indicated as mean influx rate, K1 [mg Fe/d], into the blood pool accumulates in the body iron store, U s ( t ) . Minor influences were neglected, e.g. increasing blood pool during growth, delay of iron influx due to finite life-time of donor erythrocytes, and iron reflux from erythropoiesis. The chelation of iron by the oral chelator DFP (or by DFO) shows up as iron excretion rate constant, k2 [d-l]. In 6rst approximation, the body iron store is described by a linear differential equation, dU,/dt = K1 - k2 - Us. This equation is solved analytically by the function (2).

At least square fit of this function to the experimental data (body iron stores, Us(t) , and transfused iron, K1) results in a mean total iron excretion rate constant, k2. Initial body iron stores, Uo = LJ,(t,), were measured before a period of therapy. For the time of biomagnetometry, the total body

RESULTS

Liver iron concentrations and serum ferritin values were determined simultaneously in 52 patients with secondary iron overload. Only a poor correlation between these parameters was observed (r2 = 0-49, n = 57) as shown in Fig 2. Regarding these two parameters which are used to describe the range of the respective iron overload in a given patient, no difference was obvious between groups of patients who were on long-term iron chelation ( > 2 years) either with DFO or DFP (Fig 2).

12000 A A

0 3 6 9 12 15

liver iron [mg/g,iver]

Fig 2. Serum ferritin versus liver iron concentration in patients with secondary iron overload. Liver iron was measured by SQLJID biomagnetic susceptometry. Crosses represent patients under treatment with DFP (thin crosses for 0.2-2 years: bold crosses for 2 - 5 . 5 years.

The Iower quartile of liver siderosis was found to be <1.7 mg/gliver, whereas the upper quartile was > 4 5 mg/gkver. In the interquartile range (SO%), serum ferritin values ranged from 530 to 7793 ,ug/l, with a median value of 2477pg/l. The broad distribution of serum ferritin values indicates the limited use of individual serum ferritin to also quantify iron stores in this most relevant group of patients.

The iron metabolism in six patients under DFP treatment for 3-15 months (Table I) was studied in more detail. The efficacy of iron excretion was measured by changes in liver (and spleen) iron concentration using biomagnetic suscept- ometry. A single compartment model function (eq. 2 , Fig 1) was applied to the experimental data from liver and spleen. These patients were under regular blood transfusion therapy

0 1995 Blackwell Science Ltd. British Journal of HQernQtolOg~ 91: 827-833

830 P . Nielsen et a1 (transfused iron 12-55mg/d, Table 11) and had also a history of iron chelation treatment with DFO. At the beginning of the DFP therapy, liver iron concentrations ranged between 4 and 10mg/g liver (Table I). Values of spleen iron concentrations were 50% lower than the respective liver values. Assuming 90% of the excessive storage iron in the liver and spleen in iron loaded patients, the initial total body iron stores were in the range of 5-36 g (Table I).

During the DFP-treatment period, the iron store decreased in most patients as shown in Table II and in Fig 3. However, single serum femtin values varied strongly during iron chelation therapy. In order to get a more representative trend of the ferritin data in the whole DFP interval, a linear figure was calculated with the regression coefficients shown in Table IT. Although a negative iron balance was indicated

correctly by most of these values, no significant correlation was found due to the large variation coefficients of the intra- individual serum ferritin trend values (36% to >1000/, Table 11).

Urine iron excretion under DFP treatment varied between 2 mg/d (patient 6) and 100 mg/d (patient 3) with mean rates out of at least three collection periods shown in Table 11. Complete collection was checked by 4')K measurements of the urine samples. Normalization of urine iron to assumed normal potassium excretion would result in larger iron excretion in the case of incomplete urine collection (Table II, patients 4-6). A contradictory result was found in patient 3 regarding large urine iron excretion, but rather unchanged values of liver iron concentration and serum ferritin. The discrepancy would even be increased if correction for normal potassium excretion was to be discarded because of increased

'lo$ :i 4

- I

C .- &d - .- L L

.c a - E 2 8

-1 03

+ + +

+

+ +++ h +

2i 0 21 00 2400 2700 3000 3300 3600

t ++ +

Pig 3. Negative iron balance in two patients with Pthalassaemla major under transfusion therapy with oral *prone @m) or S.C. dehxamine @PO). (a) Patient 4 (see Tables I and II); (b) patient 5 (see Tables I and II).

+ Squares: iron stores MD. calculated from measured values of Uver and spleen iron. Crosses: serum ferritin values. Curve: applied

o ~ l ~ l " l " l ' ~ l ~ '

1500 1800 21 00 2400 2700 3000 model fundon with mean iron excretion rate constant. a 9 8 [dl

0 1995 Blackwell Science Ltd, Brftfsh Journal of Haernatologu 91: 827-833

Biomagnetometrg in Iron Chelation Therapy 83 1 marrow. However, the invasive nature of these procedures limits their clinical usefulness, especially when repeated measurements are needed for monitoring the efficacy of iron depletion therapy. Recent development in physical methods such as computed tomography (CT), magnetic resonance imaging (MRI), and biomagnetic liver susceptometry (BLS) were used to measure liver iron concentration in humans non-invasively. Among this, single- and dual-energy CT proved to have a too low sensitivity in measuring liver iron (Guyader et al, 1989; Nielsen et d. 1992). In addition, repeated CT measurements would result in a substantial radiation burden, especially in children.

Quantitative MRI has been demonstrated to measure liver iron precisely in patients with hereditary haemochromatosis (Kaltwasser et al. 1990 Engelhardt et al, 1994). However, this method is restricted to medium-grade liver siderosis. Higher liver iron concentration, as trpically found in patients with transfusional iron overload, can hardly be quantiied with the MRI equipment presently available (Engelhardt et al, 1994). This restriction is not valid for biomagnetic susceptometry. The method of BLS has been confirmed in more than 50 patients with haemochromatosis or transfusional siderosis undergoing one or two liver biopsies (Fischer et ul, 1992). As the method is based on a physical calibration only and as long as magnetic suscept- ibilities for liver and spleen ferritin/haemosiderin do not differ, the method can be applied also to the quantification of spleen iron. It was shown earlier (Brittenham et al, 1982, 1993), and in the present investigation, that BLS seems to be the most suitable method to measure liver iron precisely and non-invasively .

The non-invasive measurement of liver (spleen) iron offers some advantages in the therapy control of patients under iron chelation therapy. First, detailed and reliable informa- tion on the iron balance is available. This enables a more rational use of different iron chelators in the future and the adaptation of optimal dosage of a given medication. Iron balance in a certain therapy interval can be measured directly as change of iron stores and has to be the difference between the transfused iron (+ the unknown amount of nutritional iron) and the totally excreted iron (see eq. 3). As can be seen from Table 11, there are significant discrepancies between total body and urinary iron excretion rates. Although the quality of urine data is relatively bad due to incompliance and short collection periods of 24-48 h during different therapy intervals, there seems to be a lack of urine iron between 20% and 50%. As faecal iron excretion does not contribute significantly to total body excretion (Olivieri et al. 1990). other explanations have to be considered for the missing amount of iron in most patients.

A second point concerns the compliance of patients under iron chelation therapy, which is a known problem in the long-term treatment of patients with /3-thalassaemia. More direct information on the degree of iron loading can help to increase the acceptance of an effective iron chelation therapy. In addition, the effective compliance of a patient can be judged more objectively. An example of a suspected low compliance is demonstrated in patient 3 (Tables I and 11). Under DFP treatment, large urinary iron excretion with

potassium turnover. A low compliance for iron chelation therapy is suspected in this patient.

For all six patients, the two-compartment model function (eq. 2) was applied to the experimental data (liver and spleen iron). Examples of descriptions of the body iron stores under iron chelator treatment are outlined in Fig 3(a) (patient 4) and Fig 3(b) (patient 5 ) . Mean iron excretion rate constants ( k 2 = 0.0033 f 0.0003 d-' and k2 = 0 ~ 0 0 4 1 ~ 0 ~ 0 0 0 1 d-l) were fitted to the data of patients 4 and 5 during DFP as well as DFO intervals. A significant deviation from the mean model curve during each of the three DFO and DFP intervals would indicate a different effectiveness of both iron-chelating compounds. Intra-individual mean specific rate constants (100 - k 2 / K ~ , D F p ) in the three intervals were

0.23 f o-o8%/g,,FO, 0.23 f 0'08%/8,, for patients 4 and 5, respectively.

Mean total body iron excretion rates, FemEwon, in the relevant DFP interval were calculated according to equation 3 (see Table II) from the difference of transfused iron and iron storage balance, i.e. Fe,xcr,won = Fetransfusion - Febalanee. A clear negative iron storage balance (mean fSEM: 6-5 rfI 1.7mg/d) could be demonstrated by the model calculation as well as by the direct experimental biomagnetic data (Fig 3. Table 11). The positive iron balance (Febalance) for patient 1 was explained by the very high iron transfusion rate of 55 mg/d (-8 erythrocyte concentrates per month).

found aS 0.20 f O*11%/ gDF0, 0.17 f 0.03%/8Dpp and

DISCUSSION

The liver (and spleen) iron concentrations were measured non-invasively by biomagnetic susceptometry in six patients with severe iron overload under iron depletion therapy with the oral iron chelator deferiprone. Together with the sonographic determination of the organ volumes, the actual body iron stores could be determined more accurately in all patients. A simple exponential model function was applied to the experimental data, resulting in a detailed description of the iron metabolism. Based on these data, a clearly negative iron storage balance could be demonstrated in five out of six patients under treatment with DFP. This was also demonstrated in 21 patients in a recent prospective study by Olivieri et al (1995) using BLS in a similar way. From their results one could estimate a negative liver iron balance of roughly 7 f 2 mg/d for their group of previously ineffectively chelated patients.

Throughout iron chelation therapy with DFP, the intra- individual serum femth values showed a considerable fluctuation and gave imprecise information on the progress of therapy. This c o n h e d former studies in iron-loaded patients which showed a limited use of serum ferritin measurements to (i) estimate on the individual body iron stores in patients with /3-thalassaemia (Chapman et al, 1982; Brittenham et al, 1993) or (ii) to monitor iron mobilization under conditions where inflammation or tissue damage is present (Worwood et aI, 1980; Kaltwasser et aZ, 1989).

Detailed information on the amount and distribution of tissue iron overload can be obtained from biopsy specimens of the major iron storage organs such as liver and bone

0 1995 Blackwell Science Ltd, British Journal of Haemutology 91: 827-833

832 P. Nielsen et a1 maximum values of 11 6 mg Fe/d during repeated observa- tion intervals were measured, whereas the iron balance and the serum ferritin trend values (Table II) were only slightly negative. These data can only be matched if large extra- hepatic iron stores (>13 g) are present and/or a nutritional iron absorption rate of up to 33 mg/d is assumed. Although there is some evidence that excess storage iron in this thalassaemic patient is partly stored in muscle, bone marrow and skin, the amount of this iron pool will not be comparable with the liver iron. Consequently, an only occasional intake of DFP preferentially before urine collection periods is suspected in this case as the only plausible reason for the observed discrepancies. It should be pointed out that this information could not be derived from fluctuating changes in serum ferritin values or from the urinary iron excretion data.

Several conclusions can be drawn from the model considerations according to equation (2). Assuming regular transfusion therapy as well as an efficient and constant iron chelation treatment in patients 4 and 5 (Fig 3 ) , a steady- state condition will be obtained and the iron stores will be represented by the plateau values of Kl/kz after about 3 years. Local significant deviations of the experimental data from this ideal behaviour are to be seen in Fig 3 . These deviations are largely explained by differing transfusion and chelation rates during certain therapy intervals, e.g. 6rst DFO interval for patient 5 in Fig 3(b) with only 0*7gDF0/d and k2 = 0.0017d-1 had to be separated from the more comprehensive model calculation. However, in view of a limited chelatable iron pool one could also discuss a non- linear behaviour of the excretion rate constant, k 2 , with iron stores. Indeed, there is some indication for this hypothesis towards smaller iron concentrations in Fig 3 for those two patients who have been monitored from initial severe iron loading to average liver iron concentrations adequate for most of these patients. Thus, it becomes also obvious that an intra-individual comparison between similar effective iron- chelating compounds (DFO and DFP) is difficult to perform in patients under continuous therapy within a therapy period even when repeated liver iron quantification is available. This may become easier for those patients in whom equilibrium between transfusion and chelation therapy has already been achieved.

Concerning the efficacy of DFP, it has been shown in the past that various doses (25-lllmg/kg/d) can induce a trend towards decreased serum ferritin values in most patients (Olivieri et al, 1990; Tondury et al, 1990; Al- Refaie et 111, 1992; Agarwal et ul, 1992). In addition, the urinary iron excretion was found similar under DFP treatment as compared to equivalent doses of DFO.

However, every new oral iron chelator has to demonstrate its capability to really deplete excessive iron stores in iron- loaded patients, as has been shown for deferoxamine (Cohen et al, 1984) and recently also by BLS in 21 patients receiving deferiprone for about 3 years (Olivieri et al, 1995). As serum ferritin is a simple but only qualitative parameter, repeated non-invasive liver iron quantification, preferentially in intervals of 1-2 years using the SQUID biomagnetometry technique, is the method of choice to monitor the efficacy of iron chelation therapy.

ACKNOWLEDGMENTS

We thank Susanne Hoppe and Rosemarie Kongi for their excellent technical assistance and, in particular, Christian Tiemann for taking care of patients and performing measurements.

REFERENCES

Agarwal, M.B., Gupte, S.S., Viswanathan, C., Vasandani, D., Ramanathan, J., Neena Desai. Puniyani, R.R. & Chhablani. A.T. (1992) Long-term assessment of efficacy and safety of L 1 , an oral iron chelator, in transfusion dependent thalassaemia: Indian trial. British Journal of Haematology. 82.460-466.

Al-Refaie, F.N., Wonke, B., Hoffbrand, A.V., Wickens, D.G., Nortey, P. & Kontoghiorghes, G.J. (1992) Efficacy and possible adverse effects of the oral iron chelator 1,2-dimethyl-3-hydroxypyrid-4- one (Ll) in thalassemia major. Blood, 80. 593-599.

Brittenham, G.M., Cohen, A.R., McLaren, C.E., Martin, M.B., Grifflth, P.M.. Nienhuis, A.W., Young, N.S., Allen, C.J., Farrell. F. &Harris, J.W. (1993) Hepatic iron stores and plasma ferritin concentration in patients with sickle cell anemia and thalassemia major. American Journal of Hematology. 42, 81-85.

Brittenham, G.M., Farrell, D.E.. Harris, J.W., Feldman, E.S., Danish, E.H.. Muir, W.A., Tripp, J.H. & Bellon, E.M. (1982) Magnetic susceptibility measurement of human iron stores. New England Journal of Medicine, 307. 1671-1675.

Chapman, R.W., Hussain. M. & Gorman. A. (1982) Effect of ascorbic acid deficiency on serum ferritin concentration in patients with beta-thalassemia major and iron overload. Journal of Clinical

Cohen. A,, Martin, M. & Schwarz. E. (1984) Depletion of excessive liver iron stores with desferroxamine. British Journal of Haematologg, 58, 369-373.

Engelhardt, R., Langkowski, J.H.. Fischer, R., Neilsen, P., Kooijman, H., Heinrich, H.C. & Bucheler, E. (1994) MRI-sequences for liver iron quantification: studies in aqueous iron solutions, iron overloaded rats and patients with hereditary hemochromatosis. Magnetic Resonance Imaging. 12. 999-1007.

Fischer, R.. Eich, E., Engelhardt, R.. Heinrich. H.C., Kessler, M. & Nielsen, P. (1989) The calibration problem in liver iron susceptometry. Advances in Biomagnetism (ed. by S. J. Williamson et al). pp. 501-504. Plenum Press, New York.

Fischer, R., Engelhardt, R., Nielsen. P., Gabbe, E.E., Heinrich, H.C., Schmiegel, W.H. & Wurbs, D. (1992) Liver iron quantification in the diagnosis and therapy control of iron overload patients. Advances in Biornagnetism '91: Clinical Aspects (ed. by M. Hoke et al). pp. 585-588. Elsevier, Amsterdam.

Guyader, D., Gandon, Y. & Deugnier, Y. (1989) Evaluation of computed tomography in the assessment of liver iron overload: a study of 46 cases of idiopathic hemochromatosis. (hstroenterology. 97.737-743.

Hider, R.C., Kontoghiorges, G . & Silver, J. (1982) UK Patent. GB- 2 1181 76.

Kaltwasser, J.P., Gottschalk. R., Schak. K.P. & Hartl. W. (1990) Non-invasive quantitation of liver iron-overload by magnetic resonance imaging. British Journal of Haernatologu, 74, 360-363.

Kaltwasser, J.P. & Werner. E. (1989) Diagnosis and clinical evaluation of iron overload. Bailliere's Clinical Hacmatology. 2 ,

Kontoghiorghes, G.J., Aldouri, M.A., Sheppard, 1,. & Hofirand, A.V. (198 7a) 1,2-Dimethyl-3-hydroxypyrid-4-one, an orally active chelator for treatment of iron overload. Lancet. i, 1294-1295.

Kontoghiorghes, G.J., Bartlett, A.N.. Hofirand. A.V., Goddard, J.G.. Sheppard, L.. Barr. J.. Nortey. P. (1990) Long-term trial with the

Pathologg, 35, 487-491.

363-389.

0 1995 Blackwell Science Ltd. British 7ournal of Haematology 91: 827-833

Biomagnetometry in Iron Chelation Therap3 83 3 I., St Louis, P.. Freedman, M.H., McClelland, R.A. & Templeton, D.M. (1990) Comparison of oral chelator L 1 and desferrioxamine in iron-loaded patients. Lancet. 336, 1275-1279.

Pdulson, D.N.. Fagdly. R.L., Toussant. R.M. & Fischer, R. (1991) Biomagnetic susceptometer with SQUID instrumentation. IEEE Transactions: Mngnetism, MAG-27, 3249-32 52.

Pippard, M.J., Johnson, D.K., Callender, S.T. & Finch, C.A. (1982) Ferrioxamine excretion in iron loaded man. Blood, 60, 288-294.

Shoden, A. &Sturgeon, P. (1960) Haemosiderin. Acta Haernatologica,

Tondury, P.. Kontoghlorghes, G.J.. Ridolfi-Luthy, A.. Hirt, A., Hoflbrand, A.V., Lottenbach, A.M., Sonderegger, T. & Wagner, H.P. (1990) Ll(1.2-dimethy1-3-hydroxypyrid-4-one) for oral iron chelation in patients with beta-thalassaemia major. British Journal of Haematology, 76. 550-553.

Wonvood, M.. Cragg, S.J., Jacobs, A., McLaren, C.. Ricketts, C. & Econmidou, J. (1 980) Binding of serum ferritin to concanavalin A: patients with homxygous /3 thalassaemia and transfusional iron overload. British Journal of Haematology, 46, 409-416.

23. 376-392.

oral iron chelator 1,2-dimethyl-3-hydroxy-pyrid-4-one. I. Iron chelation and metabolic studies. British Journal of Haernatology,

Kontoghiorghes. G.J. & Sheppard, L. (1987b) Simple synthesis of the potent iron chelators l-akyl-3-hydroxy-2-methylpyrid4-ones. lnorganica Chimica Acta. 136, Lll-L12.

Leung. N.W.Y., Farrant. P. & Peters, T.J. (1986) Liver volume measurement by ultrasonography in normal subjects and alcoholic patients. Journal of Hepatology, 2, 157-164.

Nielsen, P., Engelhardt. R., Fischer, R., Heinrich, H.C., Langkowski. J.H. & Biicheler, E. (1992) Noninvasive liver-iron quantification by computed tomography in iron-overloaded rats. Investigative Radiology. 27, 312-317.

Olivieri, N.F., Brittenham, G.M., Matsui, D.. Berkovitch, M., Blendis, L.M.. Cameron, R.G.. McClelland, R.A., Liu. P.P., Templeton. D.M. & Koren, G. (1995) Iron-chelation therapy with oral deferiprone in patients with thalassemia major. New England Journal of Medicine, 332, 918-922.

Olivieri, N.F., Koren, G., Hermann, C., Bentur, Y., Chung, D., Klein,

76, 295-300.

0 1995 Blackwell Science Ltd, British Journal of Huematology 91: 827-833