Mortality for patients hospitalized for cardiac arrest is still very high. In France, as in Europe, population is growing older and management of cardiac arrest in the elderly raises medical and ethical questions.

To compare one-year mortality according to the age (<70 years, ≥70 years) for patients admitted for cardiac arrest, and to describe predictors of survival in the group ≥70 years.

All patients admitted for non-traumatic cardiac arrest in our center from 1 January 2015 to 31 December 2016 were retrospectively included. Patients were analyzed depending on the age: under and over 70 years old.

Two hundred and ninety one patients were included (203 under 70 years, 88 over 70 years), whose 38 for in-hospital cardiac arrest. In-hospital mortality (138 (67.9%) vs. 71 (80.7%); P=0.27), one-month (140 (68.9%) vs. 74 (84.1%); P=0.001) and one-year mortality (144 (70.9%) vs. 78 (88.6%); P=0.001) were significantly higher in the group over 70 years. In this group, initial shockable rhythm (OR 0.25; IC 95% 0.05-0.73) and a medical or para-medical witness (OR 0.28; IC 95% 0.09–0.83) were associated with a good prognosis whereas diabetes (OR 8.7; IC 95% 1.08–69.7), cardiac arrest at home (OR 3.5; IC 95% 1.1–10.8), longer low-flow (OR 1.2; IC 95% 1.08–1.4) and higher lactates (OR 1.22; IC 95% 1.04–1.43) were associated with lower survival to hospital discharge.

Older age is associated with a higher mortality after a cardiac arrest. In patients under 70 y.o, initial shockable rhythm and medical or para-medical witness were associated with a better prognosis, whereas diabetes, cardiac arrest at home, a longer low-flow and higher lactates were associated with a poor prognosis.