Importance With increasing altitude, the arterial partial pressure of oxygen (PaO₂) decreases. However, the extent of the reduction as a function of altitude remains unknown. The aim of this study was to present an effect size for the decrease in PaO₂ that comes with each kilometre vertical gain and to identify factors influencing PaO₂ at altitude.

Methods A systematic search of PubMed and Embase was performed. Peer-reviewed, prospective studies in healthy adults providing arterial blood gas analysis at low altitude (<1500m) and within the first 3 days at the target altitude (?1500m) were analysed. Data were extracted from the included studies and if possible, the individual patient data (IPD) was obtained. Main outcomes were mean estimates and 95% confidence intervals of the association between PaO₂ and altitude in healthy adults.

Results 53 studies (777 subjects, 34.4% female) spanning an altitude range from 1524m to 8730m were included in the aggregated data analysis and 13 studies (305 subjects, 45.1% female) were included in the IPD analysis. The estimated effect size for PaO₂ was -1.60kPa [-1.73 to -1.47kPa] for each 1000m of altitude gain. Baseline predictors for PaO₂ at altitude were target altitude (coef. -1.52kPa/km), age (-0.013kPa/y) and days ?1500m (0.157kPa/day), whereas sex, BMI and baseline PaO2 were not.

Conclusion This meta-analysis provides estimates of altitude-related reductions in PaO₂ in healthy individuals above 1500m. This effect estimate of 1.60kPa/1000m vertical gain for healthy people will provide a better understanding of hypobaric hypoxia in healthy individuals and a basis for investigation chronically ill people at altitude.