Thirteen healthy European mountaineers (11 male, 2 female) participated in the 62-day German-Pakistani Research Expedition to Broad Peak (8047 m) in the Karakorum, Pakistan. During ascent, base camp stay and approach to the summit, oxygen saturation was measured by pulse oximetry at rest, during exercise and during sleep; in addition, questionnaires on high altitude symptomatic had to be answered. We found a dramatic decrease in oxygen saturation especially at extreme altitudes (7100 m: Median 63%, Min 59%, Max 65%) and a long time required for real acclimatization. The lowest figures at 4850 m were found during maximal exercise, 77.5% (69 - 85%) and during sleep, 81% (73 - 88%), the highest ones at rest, 86.5% (77 - 89%). There was a significant correlation (Spearman rank correlation coefficient with ties) between measured oxygen saturation during the ascent to/stay at base camp and high altitude illness (p = 0.005 - 0.05), as well as with high altitude performance (p = 0.025 - 0.01). The limiting values of "no high altitude symptomatic", "high altitude discomfort", AMS and the malignant forms could be estimated for acclimatized (>90%/>80%/>70%/<70%) and unacclimatized (>80%/>70%/>65%/<65%) condition. Pulse oximetry is an objective non-invasive method of measurement that is easy to handle. It is a suitable device besides clinical examination and questionnaire-test in the diagnosis of high altitude illness even in the hands of non-professionals. The measurement at sleep can possibly explain present high altitude symptomatic despite of (nearly) normal oxygen saturation values at rest.