Whichever one you want. By numerical fatigue calculation standards, they are essentially the same result. You always should consider fatigue results in a statistical sense. There is never one correct answer. There is a range of correct answers depending on factors such as how comprehensive the fatigue test data is, how well known the applied loading is, how precisely the surface effects are known, etc. Most commercial design codes apply a large factor of safety on the calculated fatigue life to encompass the known unknowns and the inherent statistical variance.
EN is generally considered to be more mathematically precise than SN, but often the SN test data is more comprehensive, so it can be more accurate.
You are navigating away from the AIS Discovery experience