Abstract 1024


We previously established an association between increasing liver iron concentration (LIC) and vascular as well as endocrine/bone disease in patients with β-thalassemia intermedia (TI) (Musallam et al. Haematologica 2011). More recently, a randomized clinical trial showed the efficacy and safety of deferasirox for the treatment of iron overload in TI patients with a LIC ≥5 mg Fe/g dry weight (dw) (Taher et al. Blood 2012). The aim of this analysis was to evaluate the association of the 5 mg Fe/g dw threshold with vascular and endocrine/bone morbidity in TI patients.


We analyzed data from a cross-sectional study of 168 TI patients who never received iron chelation therapy. For each patient, iron burden was determined directly by measuring LIC using magnetic resonance imaging. Data on the occurrence of vascular (thrombosis or pulmonary hypertension) as well as endocrine/bone (hypothyroidism, osteoporosis, or hypogonadism) morbidities were also retrieved. Splenectomy status and transfusion history, as well as total hemoglobin level at the time of LIC measurement were also determined.


The mean age of patients was 35.2 ± 12.6 years (range: 8–66 years) with 42.9% being males and 72.0% being splenectomized. The mean total hemoglobin level was 8.8 ± 1.6 g/dl, with 26.2% being completely transfusion-naive while the remaining patients had received some form of transfusion therapy for specific complications and for defined periods of time. The mean LIC was 8.4 ± 6.7 mg Fe/g dw (range: 0.5–32.1 mg Fe/g dw), with 70 (41.7%) patients having a LIC of <5 mg Fe/g dw and 98 (58.3%) having a LIC of ≥5 mg Fe/g dw. A total of 35 (50.0%) patients had at least one morbidity in the <5 mg Fe/g dw group compared with 83 (84.7%) in the ≥5 mg Fe/g dw group (p<0.001). The absolute morbidity risk increase attributable to the ≥5 vs.<5 mg Fe/g dw threshold (84.7% minus 50.0% = 34.7%) was as high as that attributed to the ≥7 vs. <7 mg Fe/g dw threshold (88.6% minus 53.9% = 34.7%); the latter being a historical prognostic threshold of increased morbidity in patients with β-thalassemia major. The ≥5 mg Fe/g dw group had a significantly higher prevalence of all evaluated morbidities compared with the <5 mg Fe/g dw group (thrombosis: 34.7% vs. 14.3%, p<0.01; pulmonary hypertension: 43.9% vs. 18.6%, p<0.01; hypothyroidism: 24.5% vs. 8.6%, p<0.01; osteoporosis: 58.2% vs. 28.6%, p<0.001; and hypogonadism: 23.5% vs. 7.1%, p<0.01). The prevalence of patients with multiple morbidities was also higher in the ≥5 than the <5 mg Fe/g dw group (60.2% vs. 17.1%, p<0.001). To determine whether the observed association between LIC and morbidity is confounded by clinically relevant risk factors, we adjusted the association for age, sex, total hemoglobin level, splenectomy, and transfusion using logistic regression analysis. The unadjusted effect estimate (odds ratio) for the ≥5 vs. <5 mg Fe/g dw threshold with morbidity as the dependent variable was 5.53 (95% CI: 2.69–11.40). Upon adjustment, the odds ratio dropped minimally to 3.76 (95% CI: 1.62–8.71) indicating that the observed association is primarily independent of such confounders.


A LIC of ≥5 mg Fe/g dw is independently associated with a considerably increased risk of vascular and endocrine/bone disease in patients with TI. Administering iron chelation therapy for patients exceeding this threshold is thus not only expected to lower LIC but may be associated with a reduction in the risk of serious morbidities that are often irreversible.


Musallam:Novartis Pharmaceuticals: Honoraria. Cappellini:Novartis Pharmaceuticals: Speakers Bureau. Taher:Novartis: Honoraria, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.