Risk factors for heart disease in transfusion-dependent thalassemia: serum ferritin revisited

• DOI: 10.1007/s11739-018-1890-2
• Publication Date: 2018-06-12T14:10:08Z
• Authors: Giorgio Derchi; Carlo Dessì; Patrizio Bina; Maria Domenica Ca...; Antonio Piga; Silverio Perrotta; Immacolata Tartag...; Marianna Giuditta; Filomena Longo; Raffaella Origa; Antonella Quarta; Valeria Pinto; Gian Luca Forni
• Keywords: Adult; Male; Chi-Square Distribution; Heart disease; Iron overload; Serum ferritin; Thalassemia; Adolescent; Adult; Area Under Curve; Blood Transfusion; Chi-Square Distribution; Child; Cohort Studies; Female; Ferritins; Heart Diseases; Humans; Italy; Male; ROC Curve; Retrospective Studies; Risk Factors; Survival Analysis; Thalassemia; Adolescent; Heart Diseases; Survival Analysis; 3. Good health; Cohort Studies; 03 medical and health sciences; 0302 clinical medicine; Italy; ROC Curve; Risk Factors; Area Under Curve; Ferritins; Humans; Thalassemia; Blood Transfusion; Female; Child; heart disease; iron overload; Serum ferritin; Thalassemia; Internal Medicine; Emergency Medicine; Retrospective Studies

Heart disease remains a leading cause of morbidity and mortality in transfusion-dependent thalassemia (TDT), which can be attributed to several factors but primarily develops in the setting of iron overload. This was a retrospective cohort study utilizing Webthal® patient data from five major centers across Italy. Patients without heart disease were followed-up for 10 years (2000-2010) and data were collected for demographics, splenectomy status, serum ferritin and hemoglobin levels, and comorbidities associated with heart disease. Among 379 patients analyzed (mean age 22.9 ± 5.1 years, 47.8% men), 44 (cumulative incidence: 11.6%) developed heart disease during the period of observation. Splenectomy (p = 0.002) and serum ferritin level (p < 0.001) were the only risk factors with significant association with heart disease. A serum ferritin threshold of ≥ 3000 ng/mL was the best predictor for the development of heart disease (86.4% sensitivity and 92.8% specificity, AUC: 0.912, 95% CI 0.852-0.971, p < 0.001). On multivariate analysis, only a serum ferritin level ≥ 3000 ng/mL remained significantly and independently associated with increased risk of heart disease (HR: 44.85, 95% CI 18.85-106.74), with a 5- and 10-year heart disease-free survival of 58 and 39%. The association between iron overload and heart disease in patients with TDT is confirmed, yet a new serum ferritin level of 3000 ng/mL to flag increased risk is suggested.