Introduction: Myelodysplastic syndromes (MDS) and aplastic anemia (AA) are the most common anemias that require transfusion therapy in Japanese patients. The goal of this retrospective survey was to investigate the relationship between iron overload, chelation practices and morbidity/mortality in these patients.

Methods: This retrospective survey investigated the outcome of iron overload-related morbidity and mortality in Japan from Aug 2001 to Dec 2005. The medical chart histories of transfusion-dependent patients (defined as >2 red blood cell [RBC] units/month for ≥6 months) were assessed via questionnaire at three time points:

  1. transfusion onset,

  2. chelation onset,

  3. study end.

Data categories included age, underlying disease, number of RBC units received (in Japan, 1 unit=200 mL), laboratory blood tests (including serum ferritin), cardiac tests, liver MRI and biopsy.

Results: In total, 43 hospitals and 292 patients participated (some patients had overlapping conditions); MDS, 52.4%; AA, 31.0%; pure red cell aplasia, 5.2%; myelofibrosis, 4.5%; other, 9.0%. Lifetime transfusions over a mean duration of 32.0 months were: ≤40 units, 19.6%; 40–159 units, 47.8%; ≥160 units 32.5%. Patients received a mean of 68.2 units during the previous year. 43.4% of patients had previously received deferoxamine (DFO) therapy; 56.6% were chelator-naïve. Of the DFO-treated patients, 53.8% had received ≤40 units and 46.2% had received >40 units. Only 8.6% received daily/continuous DFO, with the remainder receiving intermittent DFO (once per 1.9 weeks) or other less effective regimens.

The proportion of patients with abnormal laboratory parameters increased after the initiation of transfusions: SGOT ≥36 mU/mL (16.8 to 41.8%), SGPT ≥46 mU/mL (16.4 to 36.8%) and fasting blood glucose ≥121 mg/dL (39.1 to 54%). These increases correlated with transfusion history and increased ferritin levels (1673 to 4378 ng/mL), but not with age or underlying disease. In patients with these abnormal laboratory results, >90% had ferritin levels >1000 ng/mL.

Of patients in whom cardiac and liver function were evaluated, abnormalities were observed in 22% (14/64) and 73% (11/15), respectively. Patients who received DFO were those at greatest risk of cardiac dysfunction. In patients receiving continuous DFO, ferritin, SGOT, SGPT and fasting blood glucose levels improved during treatment; the proportion of patients with abnormal parameters was lower in the continuous DFO group than with other DFO regimens.

In total, 75 deaths were reported; cardiac and liver failure were noted in 24.0% and 6.7% of cases. Where serum ferritin values were available, more patients died with ferritin ≥1000 ng/mL (49%) than <1000 ng/mL (1%).

Conclusions: In patients such as those with MDS and AA, iron overload induces clinical problems similar to those observed in thalassemia. This retrospective analysis shows mortality was higher in heavily iron overloaded patients, with liver and cardiac dysfunction being the primary cause. Continuous chelation therapy was effective in reducing iron burden and improving organ function. Based on these findings, it is proposed that chelation therapy be initiated according to a serum ferritin level of 1000 ng/mL.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author