We aimed to evaluate 1) the altitude where maximal heart rate (HR (max)) decreases significantly in both trained and untrained subjects in moderate acute hypoxia, and 2) if the HR (max) decrease could partly explain the drop of V.O (2max). Seventeen healthy males, nine trained endurance athletes (TS) and eight untrained individuals (US) were studied. Subjects performed incremental exercise tests at sea level and at 5 simulated altitudes (1000, 1500, 2500, 3500, 4500 meters). Power output (PO), heart rate (HR), arterial oxygen saturation (SaO (2)), oxygen uptake (V.O (2)), arterialized blood pH and lactate were measured. Both groups showed a progressive reduction in V.O (2max). The decrement in HR (max) (DeltaHR (max)) was significant from 1000 m for TS and 2500 m for US and more important in TS than US (at 1500 m and 3500 m). At maximal exercise, TS had a greater reduction in SaO (2) (DeltaSaO (2)) at each altitude. DeltaHR (max) observed in TS was correlated with DeltaSaO (2). When the two groups were pooled, simple regressions showed that DeltaV.O (2max) was correlated with both DeltaSaO (2) and DeltaHR (max). However, a multiple regression analysis demonstrated that DeltaSaO (2) alone may account for DeltaV.O (2max). Furthermore, in spite of a greater reduction in SaO (2) and HR (max) in TS, no difference was evidenced in relative DeltaV.O (2max) between groups. Thus, in moderate acute hypoxia, the reduction in SaO (2) is the primary factor to explain the drop of V.O (2max) in trained and untrained subjects.