HACE was defined by clinical signs of neurological deficit. The diagnosis of HAPE or HACE was always confirmed by an expert, either on the spot where the BMN673 PARP inhibitor disorder occurred or later on when PF-4217903 side effects hospitalized. Subjects were asked to note any medication taken during their stay, especially acetazolamide, widely used among trekkers for the prevention of AMS. From the questionnaire, subjects were classified as no or moderate AMS, severe AMS, HAPE or HACE. Severe AMS, HAPE and HACE were pooled as SHAI and considered as being intolerant to high altitude. Information about the ascent characteristics was collected in the follow-up questionnaire. This questionnaire is part of the usual care as it is systematically given to all patients coming to the consultation in order to have a follow-up in case of health problems during the stay at high altitude. History of SHAI is a major risk factor of SHAI. However, it can be determined only in subjects with previous exposure to high altitude. Therefore the entire analysis was stratified according to previous history of high altitude sojourn. The baseline characteristics of subjects were compared according to their previous history of sojourn at high altitude using Student ttest or Wilcoxon-Mann-Whitney and the Pearson chi square-test or Fisher��s exact test, as appropriate. Due to deviation from linearity assumption, physiological variables were expressed in dummy variables defined by tertiles in the group of subjects without SHAI. Two multivariate models were built for each group, one with the classical risk factors and one with supplemental adjustment for physiological variables. The variables entered in the multivariate model were those previously identified: history of SHAI, planned rapid ascent, history of migraine, geographical location, age, sex, physical activity for the classical risk factors and hypoxic ventilatory response, hypoxic cardiac response and desaturation in hypoxia at exercise for physiological variables. Two-by-two analyses were performed to assess potential interactions and confounding by fitting. We examined interactions between these variables, and between these variables and the presence of previous high altitude sojourn.