As survival and neuromuscular function in Duchenne muscular dystrophy (DMD) have improved with glucocorticoid (GC) therapy and ventilatory support, cardiac deaths are increasing. Little is known about risk factors for cardiac and non-cardiac causes of death in DMD. A multi-center retrospective cohort study of 408 males with DMD, followed from January 1, 2005 to December 31, 2015, was conducted to identify risk factors for death. Those dying of cardiac causes were compared to those dying of non-cardiac causes and to those alive at study end. There were 29 (7.1%) deaths at a median age of 19.5 (IQR: 16.9-24.6) years; 8 (27.6%) cardiac, and 21 non-cardiac. Those living were younger [14.9 (IQR: 11.0-19.1) years] than those dying of cardiac [18 (IQR 15.5-24) years, p = 0.03] and non-cardiac [19 (IQR: 16.5-23) years, p = 0.002] causes. GC use was lower for those dying of cardiac causes compared to those living [2/8 (25%) vs. 304/378 (80.4%), p = 0.001]. Last ejection fraction prior to death/study end was lower for those dying of cardiac causes compared to those living (37.5% ± 12.8 vs. 54.5% ± 10.8, p = 0.01) but not compared to those dying of non-cardiac causes (37.5% ± 12.8 vs. 41.2% ± 19.3, p = 0.58). In a large DMD cohort, approximately 30% of deaths were cardiac. Lack of GC use was associated with cardiac causes of death, while systolic dysfunction was associated with death from any cause. Further work is needed to ensure guideline adherence and to define optimal management of systolic dysfunction in males with DMD with hopes of extending survival.