Diabetes Mellitus and Cardiac Complications in Thalassemia Major Patients

In thalassemia major (TM) patients myocardial iron overload and chronic anemia are the recognized leading causes of cardiomyopathy, but a role can also be played by other factors such as endocrine abnormalities. The aim of this retrospective study was to evaluate if diabetes mellitus (DM) was associated with an higher prevalence and risk of cardiac dysfunction and of heart complications, regardless to the presence of myocardial iron overload.

From a cohort of 957 TM patients who underwent MRI within the MIOT network (Myocardial Iron Overload in Thalassemia), among the patients (N = 358) with no cardiac iron (all cardiac segments with a T2* ≥ 20 ms) we identified 29 patients with DM and 329 patients without DM. The normal values of ejection fraction (EF) normalized by sex and age, obtained in a cohort of 142 TM patients without cardiac disease and iron overload, were used to define left ventricular (LV) and right ventricular (RV) heart dysfunction (EF < mean – 2 standard deviation). Heart failure (HF) was diagnosed by Magnetic Resonance Imaging (MRI) in presence of a LV and/or RV EF lower than 4 standard deviations from the normalized mean value and by a positive history (clinical symptoms, confirmation by physical examination and treatment). Myocardial fibrosis was evaluated by delayed enhancement MRI technique.

The prevalence of overall heart dysfunction (LV, RV or both) was higher in patients with DM (44.8%) versus patients without DM (28.3%), with a P-value very close to the statistically significance (P=0.061). In more details, patients with DM presented significantly more biventricular dysfunction (20.7% vs 7.6%, P=0.016). The prevalence of myocardial fibrosis was significantly higher in the DM patients vs the no DM patients (37.5% vs 19.2 %; P=0.033). Cardiac complications occurred with a significantly higher frequency in patients with DM (55.2% vs 15.5%, P<0.0001). Taking into account each cardiac complication separately, a significant difference between the groups was found in the occurrence of heart failure (27.6% vs 9.4%, P<0.003) and hyperkinetic arrhythmias (34.5% vs 5.2%,P<0.0001), both supraventricular (27.6% vs 4%, P<0.0001) and ventricular (6.9% vs 0.6%, P=0.034).

Table 1 shows odds ratios (OR) estimating the relationship between diabetes and cardiac involvement. Among cardiac dysfunctions, only the biventricular forms were significantly positively associated with the diabetes. However, the correction for age caused the loss of the significance. The association between DM and myocardial fibrosis became not significant after the correction for age and endocrine co-morbidity. Patients with DM were significantly more likely to have cardiac complications and the results were not affected by the adjustment for age and/or endocrine co-morbidity. Considering separately each cardiac complication, a significant association was found for HF and hyperkinetic arrhythmias. The association between DM and HF resulted not significant after the correction for age. For the hyperkinetic arrhythmias, the OR remained significant after the correction for age and/or endocrine co-morbidity.

In TM patients without myocardial iron DM was significantly associated with the presence of cardiac complications globally considered and hyperkinetic arrhythmias.

No relevant conflicts of interest to declare.