PurposeBoth exercise and hypoxia cause complex changes in acid–base homeostasis. The aim of the present study was to investigate whether during intense physical exercise in normoxia and hypoxia, the modified physicochemical approach offers a better understanding of the changes in acid–base homeostasis than the traditional Henderson–Hasselbalch approach.MethodsIn this prospective, randomized, crossover trial, 19 healthy males completed an exercise test until voluntary fatigue on a bicycle ergometer on two different study days, once during normoxia and once during normobaric hypoxia (12% oxygen, equivalent to an altitude of 4500 m). Arterial blood gases were sampled during and after the exercise test and analysed according to the modified physicochemical and Henderson–Hasselbalch approach, respectively.ResultsPeak power output decreased from 287 ± 9 Watts in normoxia to 213 ± 6 Watts in hypoxia (−26%, P < 0.001). Exercise decreased arterial pH to 7.21 ± 0.01 and 7.27 ± 0.02 (P < 0.001) during normoxia and hypoxia, respectively, and increased plasma lactate to 16.8 ± 0.8 and 17.5 ± 0.9 mmol/l (P < 0.001). While the Henderson–Hasselbalch approach identified lactate as main factor responsible for the non-respiratory acidosis, the modified physicochemical approach additionally identified strong ions (i.e. plasma electrolytes, organic acid ions) and non-volatile weak acids (i.e. albumin, phosphate ion species) as important contributors.ConclusionsThe Henderson–Hasselbalch approach might serve as basis for screening acid–base disturbances, but the modified physicochemical approach offers more detailed insights into the complex changes in acid–base status during exercise in normoxia and hypoxia, respectively.Electronic supplementary materialThe online version of this article (doi:10.1007/s00421-017-3712-z) contains supplementary material, which is available to authorized users.