Evaluation of myocardial and hepatic iron loading by MRI T2* in multi-transfused patients with repeated blood loss as compared to thalassaemia major patients and controls

Dear Sirs,

Blood transfusions are necessary for survival in patients with β-thalassaemia major (TM) as well as patients with repeated blood loss through the gastrointestinal tract because of conditions such as intestinal vascular dysplasia. Although it is well documented that cardiac complications, due to left ventricular dysfunction as a consequence of iron overload, are the main cause of death in TM, there are only few data about iron overload in patients who receive multiple transfusions because of repeated gastrointestinal blood loss (MT-BL)¹. The aim of this study was to evaluate, by magnetic resonance imaging (MRI), myocardial and hepatic iron deposition and cardiac function in asymptomatic populations of TM patients and MT-BL patients.

Twenty MT-BL patients, 20 TM patients and 20 controls were studied: myocardial and hepatic T2* relaxation times and right and left ventricular parameters were evaluated by 1.5 T. None of the subjects had clinical signs of heart failure according to the Framingham study criteria and the New York Heart Association (NYHA) criteria¹. Iron chelation therapy with desferrioxamine had been started in the TM patients before the age 7 years, at a daily dose of 30–50 mg/kg, given 5–6 times weekly. Chelation treatment was monitored in order to keep serum ferritin at the lowest possible levels. A serum ferritin level below 2000 ng/mL was considered as an indication of successful treatment with desferrioxamine². All TM patients were transfused every 2–3 weeks, maintaining haemoglobin levels at 12.04±0.86 g/dL. The TM patients had received a mean of 260 units of blood each (range, 210–380) prior to the time of the MRI examination. The MT-BL patients had received occasional blood transfusions and had maintained their haemoglobin at a mean level of 8.84±0.48 g/dL. The MT-BL patients had received a mean of 15 units of blood each (range, 10–35) prior to the time of the MRI examination. Blood transfusions had been given to all these patients in the 2 years preceding the study, at a frequency of 2–3 times per year. None of the MT-BL patients had received chelation treatment.

For both myocardial and hepatic T2* determinations, we followed the recent Brompton protocol¹, based on a single-breath multi-echo fast gradient-echo sequence. Imaging parameters were: TR=25.6 msec and a TE range of 2.2–22.6 msec (10 echoes). For patients with heavy liver overload (T2*<2 msec), as screened by the previously used sequence, a single-breath single-echo, fast gradient echo pulse sequence was used, with a TE range of 0, 93–2 msec. The standard cine SSFP short axis was used to evaluate the right and left ventricular ejection fractions of each patient³. Values are expressed as mean±standard deviation (SD). The statistical significance of any differences between values was determined using unpaired two-tailed Student’s t-tests. Pearson’s correlation coefficient was calculated to determine correlations between variables. p values less than 0.05 were considered statistically significant.

Haemoglobin levels were significantly lower in the MT-BL patients than in the TM patients: 8.84±0.48 versus 12.04±0.86 g/dL, respectively; p<0.001. The number of transfusions was also significantly lower in the MT-BL patients than in the TM patients: 15 (range, 10–35) versus 260 (range, 210–380), respectively; p<0.001. Likewise, ferritin levels were significantly lower in the MT-BL patients than in the TM patients, 671.6±397.9 versus 1,744±973.66 ng/mL, respectively; p<0.001 (Table I).

Table I.

Patients’ demographics and haematological parameters.

Parameters	MT-BL patients (n=20)	TM patients (n=20)	P value
Age	26±2 (20–28)	26±2 (20–28)	NS
Sex	10 F/10 M	10 F/10 M	N
Transfusions	15 (10–35)	260 (210–380)	<0.001
Hb (g/dL)	8.84±0.48	12.04±0.86	<0.001
Ferritin (ng/mL)	671.6±397.9	1,744±973.66	<0.001

Both myocardial and hepatic T2* values were significantly higher in MT-BL patients than in TM ones: 54.4±3.16 versus 15.8±3.6 msec, p<0.001 and 10.93±4.93 versus 1.41±0.5 msec, p<0.001, respectively. In control subjects, the myocardial T2* was 54±5 msec (similar to that in MT-BL patients, but significantly higher than the value in TM patients, p<0.001) while the hepatic T2* was 32±8 msec (significantly higher than the values for either MT-BL or TM patients, p<0.001) (Table II).

Table II.

Patients’ MRI characteristics.

Parameters	MT-BL patients (n=20)	TM patients (n=20)	P value
T2*liver (msec)	10.93±4.93	1.41±0.5	<0.001
T2*heart (msec)	54.4±3.16	15.8±3.6	<0.001
LVEDV (mL)	102.15±4.3	71.35±5.7	<0.001
LVESV (mL)	36.4±3.66	29.25±3.05	<0.001
LVEF (%)	64.4±2.8	58.3±2.31	<0.001
RVEDV (mL)	85±7.25	63.3±4.16	<0.001
RVESV (mL)	28.2±3.07	25±2.82	<0.001
RVEF (%)	66.9±2.5	60.85±3.048	<0.001

Left ventricular end-diastolic volume (LVEDV), left ventricular end-systolic volume (LVESV), left ventricular ejection fraction (LVEF), right ventricular end-diastolic volume (RVEDV), right ventricular end-systolic volume (RVESV) and right ventricular ejection fraction (RVEF) were all higher in MT-BL patients than in patients with TM (p<0.001). Controls had a mean LVEDV of 126.9±22.8 mL, LVESV of 39.81±11.45 mL, LVEF of 69±4%, RVEDV of 135.71±24.71 mL, RVESV of 44.48±13.27 mL and RVEF of 68±4%. These values were all higher than those in TM patients (p<0.001). Compared to MT-BL patients, controls had similar LVESV, RVESV and RVEF values. No correlation between ferritin levels and MRI parameters was documented (Table II).

In this study we evaluated MT-BL and TM patients without evidence of heart failure using MRI. We found that the TM patients had severe myocardial and hepatic iron deposition, despite their relatively low ferritin levels and apparently adequate chelation therapy. The ferritin levels did not reflect the iron status of either the myocardium or the liver. This is because ferritin reflects the total body iron overload Sand not the iron deposition in individual organs, as has already been described in previous studies^1–3. This fact further emphasises the importance of MRI in evaluating iron-overloaded patients. MT-BL patients had no evidence of myocardial iron deposition. They did, however, have hepatic siderosis, because liver is the first affected organ in the case of iron overload^1–4. Additionally, MT-BL patients had higher cardiac indices, indicative of a high output state, due to chronic anaemia. Similar changes have also been documented in other patients who receive multiple transfusions, albeit less frequently than TM patients, such as patients with thalassaemia intermedia or sickle cell disease. Hepatic siderosis in the absence of heart involvement has been reported recently in these patients^1,3.

MRI has been successfully used in the evaluation of iron overload. When iron-overloaded tissues are exposed to a magnetic field, the presence of iron causes concentration-dependent signal loss. The iron affects tissue T1, T2 and T2* relaxation times. While the T1 relaxation time decreases only moderately, T2 and T2* relaxation times decrease dramatically¹. Severe iron-overloaded patients have significantly lower myocardial and hepatic T2 and T2* relaxation times compared to those in patients with a lesser degree of iron overload^1–3.

As far as we are aware, there are limited data about the role of MRI in the evaluation of MT-BL patients. The only MRI data published on MT-BL patients are limited by the scanners, there was no evaluation of left ventricular parameters and myocardial and hepatic iron deposition were measured using T2 spin echography¹. Our findings about high cardiac output in chronic anaemias are in agreement with those in previous studies^1,3,5. The most important finding of this study is the lack of iron deposition in the heart with concurrent high cardiac output and hepatic siderosis in MT-BL patients. This is in agreement with previous studies suggesting that myocardial deposition takes place after a minimum of 75 blood transfusions^3,4.

Our findings may have important implications for patients’ treatment. Until now there have been insufficient data to support the initiation of chelation therapy in MT-BL patients. This preliminary study, showing the presence of high cardiac output without myocardial iron overload and simultaneous hepatic iron deposition, may contribute to a new approach to the treatment of patients who are multiply transfused because of chronic blood loss. However, the very small sample size and the fact that our group of patients was limited in terms of age representation are important limitations of the study. Nevertheless, this is the first time that this group has been evaluated with such a new technique and our results encourage further research with regards to diagnosis, follow-up and management of multiply transfused patients. Higher numbers of multiply transfused patients, with different causes of blood loss and/or bone marrow aplasia and different transfusion loads, should be evaluated by both MRI and liver biopsy in order to identify the exact threshold of hepatic and cardiac iron overload and the possible indication for chelation in MT-BL patients^4,5.

In conclusion, in this preliminary study, we described the imaging pattern of liver and heart in TM and MT-BL patients using MRI and documented that while iron overload plays a crucial role in both myocardial and hepatic imaging patterns in TM patients, in MT-BL patients hepatic iron overload and a high cardiac output, mimicking sickle cell disease, are the main characteristics. However, further studies are needed in order to identify the transfusion threshold capable of provoking myocardial and hepatic iron deposition and the role of chelation in MT-BL patients.