Introduction

Little is known about cardiac involvement in thalassemia intermedia (TI) using cardiovascular magnetic resonance (CMR). We investigated myocardial iron overload (MIO), biventricular parameters, and myocardial fibrosis in a large cohort of TI patients, underlying the differences between transfusion-dependent and non-transfusion-dependent patients.

Methods

We studied 252 adult TI patients (119 females, 39.5±10.4 years) enrolled in the MIOT Network. MIO was assessed using a multislice multiecho T2* approach. Biventricular function parameters were quantified by cine sequences. Myocardial fibrosis was evaluated by late gadolinium enhancement acquisitions.

Results

One-hundred and eighty-eight (74.6%) patients showed no MIO in any segment, 56 (22%) had an heterogeneous distribution (52 with global heart T2*≥20 ms), and 8 (0.3%) showed an homogeneous MIO.

Left ventricular (LV) and right ventricular (RV) dilatations were present in 113 (45%) and in 49 (19%) patients, respectively. LV dysfunction was present in the 18.0% of the cases while RV dysfunction in the 3.63%. High LV mass indexes were present in 22 (8.7%) patients.

Fifty-two/227 (22.9%) patients showed myocardial fibrosis. Myocardial fibrosis was associated to LV dysfunction (P=0.001) and high mass indexes (P=0.038).

One-hundred and fourteen patients were non-transfusion dependent (transfusion requirements absent or sporadic) while 138 patients were transfusion-dependent (regular transfusions). The mean age at start of chronic transfusions was 11.8 ± 12.3 years. Table 1 shows the comparison between the two groups. Non-transfusion-dependent patients showed significantly higher global heart T2* values and MIO with a global heart T2* < 20 ms was detected in two of them (one requiring occasional blood transfusions and one non transfused). Biventricular end-diastolic volume index, stroke volume index, left ventricular (LV) mass index, and LV cardiac index were significantly higher in the non-transfusion dependent group.

Table 1
Non-transfusion-dependentTransfusion-dependentP
Age 39.9 ± 11.5 39.2 ± 9.4 0.922 
Sex (M/F) 67/47 66/72 0.083 
Global heart T2* (ms) 38.8 ± 6.7 35.5 ± 9.2 0.014 
MIO pattern, N (%):
No MIO
Heterogeneous MIO with global T2* ≥ 20 ms
Heterogeneous MIO with global T2* < 20 ms
Homogeneous MIO 
92 (80.7)
20 (17.5)
1 (0.9)
1 (0.9) 
96 (69.6)
32 (23.2)
3 (2.2)
7 (5.1) 
0.103 
LV end-diastolic volume index (ml/m2) 99.4 ± 19.6 92.9 ± 19.1 0.009 
LV end-systolic volume index (ml/m2) 36.6 ± 11.4 34.9 ± 10.4 0.249 
LV stroke volume index 62.9 ± 12.4 58.6 ± 13.1 0.007 
LV mass index (g/m2) 69.9 ± 13.9 63.9 ± 12.9 0.004 
LV ejection fraction (%) 63.7 ± 6.8 62.5 ± 6.6 0.163 
LV cardiac index (L/min/m2) 7.6 ± 2.3 6.5 ± 2.2 0.002 
LGE, N (%) 20/105 (19) 32/122 (26.2) 0.199 
RV end-diastolic volume index (ml/m2) 92.0 ± 23.3 86.5 ± 20.8 0.048 
RV end-systolic volume index (ml/m2) 32.7 ± 14.9 31.8  ± 11.3 0.571 
RV stroke volume index 58.5 ± 14.9 54.5 ± 14.3 0.017 
RV ejection fraction (%) 64.7 ± 8.3 63.3 ± 7.5 0.168 
Non-transfusion-dependentTransfusion-dependentP
Age 39.9 ± 11.5 39.2 ± 9.4 0.922 
Sex (M/F) 67/47 66/72 0.083 
Global heart T2* (ms) 38.8 ± 6.7 35.5 ± 9.2 0.014 
MIO pattern, N (%):
No MIO
Heterogeneous MIO with global T2* ≥ 20 ms
Heterogeneous MIO with global T2* < 20 ms
Homogeneous MIO 
92 (80.7)
20 (17.5)
1 (0.9)
1 (0.9) 
96 (69.6)
32 (23.2)
3 (2.2)
7 (5.1) 
0.103 
LV end-diastolic volume index (ml/m2) 99.4 ± 19.6 92.9 ± 19.1 0.009 
LV end-systolic volume index (ml/m2) 36.6 ± 11.4 34.9 ± 10.4 0.249 
LV stroke volume index 62.9 ± 12.4 58.6 ± 13.1 0.007 
LV mass index (g/m2) 69.9 ± 13.9 63.9 ± 12.9 0.004 
LV ejection fraction (%) 63.7 ± 6.8 62.5 ± 6.6 0.163 
LV cardiac index (L/min/m2) 7.6 ± 2.3 6.5 ± 2.2 0.002 
LGE, N (%) 20/105 (19) 32/122 (26.2) 0.199 
RV end-diastolic volume index (ml/m2) 92.0 ± 23.3 86.5 ± 20.8 0.048 
RV end-systolic volume index (ml/m2) 32.7 ± 14.9 31.8  ± 11.3 0.571 
RV stroke volume index 58.5 ± 14.9 54.5 ± 14.3 0.017 
RV ejection fraction (%) 64.7 ± 8.3 63.3 ± 7.5 0.168 
Conclusions

CMR plays a key role in the management of TI patients. Heart iron (global heart T2* < 20 ms) was not common, but a quarter of the patients had some pathological segments. A consistent number of patients had the stigmata of the high cardiac output state cardiomyopathy. Myocardial fibrosis was related to the high cardiac output state. The signs of the high output state were controlled in the transfusion-dependent-patients.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution