Eur J Appl Physiol (1988) 58:298-302 EuropeanJournal of pApplied h ologs and Occupational Physiology 9 Springer-Verlag 1988 Incidence of exercise induced hypoxemia in elite endurance athletes at sea level Scott K. Powers, Stephen Dodd, John Lawler, Greg Landry, Michael Kirtley, Tipton McKnight, and Stephen Grinton Applied Physiology Laboratory-School of HPERD, Center for Exercise Science, Department of Exercise and Sport Sciences, University of Florida, Gainesville, FL, USA 32611, Louisiana State University, Baton Rouge, LA, USA 70803, Earl K. Long Hospital and Ochsner Clinic, Baton Rouge, LA, USA Summary. Recent evidence suggests that exercise- induced hypoxemia (EIH) may occur in healthy trained endurance athletes. However, at present, no data exist to describe the regularity of EIH in athletes or non-athletes. Therefore, the purpose of the present investigation was to determine the in- cidence of EIH during exercise in healthy sub- jects varying in physical fitness. Subjects (N= 68) performed an incremental cycle ergometer test to volitional fatigue with percent arterial oxyhemo- globin saturation (%SAO2) measured min-by-min. For the purpose of data analysis subjects were divided into three groups according to their level of physical training: 1) untrained (N=16), 2) moderately trained (N=27), and 3) elite highly trained endurance athletes (N=25). EIH was de- fined as a %5aO2 of _<91% during exercise. EIH did not occur in any of the untrained subjects or the moderately trained subjects. However, EIH occurred in 52% of the highly trained endurance athletes tested and was highly reproducible (r= 0.95; P< 0.05). These findings further confirm the existence of EIH in healthy highly trained en- durance athletes and suggests a rather high inci- dence of EIH in this healthy population. Hence, it is important that the clinician or physiologist per- forming exercise testing in elite endurance ath- letes recognize that EIH can and does occur in the elite endurance athlete in the absence of lung disease. Key words: Exercise -- Induced hypoxemia- In- cremental exercise -- Trained athletes -- Arterial oxyhemoglobin saturation Introduction Historically, exercise induced hypoxemia (EIH) has been thought to be exclusively mediated by Offprint requests to: S. K. Powers, Department of Exercise Science, University of Florida, Gainesville, FL 32611, USA cardiopulmonary disease. The typical textbook representation of the changes in percent arterial oxyhemoglobin saturation (SaO2) during graded exercise in healthy subjects argues that no change in SaO2 occurs until subjects reach work rates greater than 75-80% of Vo2 max. In general, SaO2 declines only 2-3% (i. e. 97% to 95%) from rest to maximal exercise in healthy individuals (Astrand and Rodahl 1986). However, there appears to be a major exception to this rule. Recent studies have confirmed the existence of a rather paradoxical EIH in healthy elite endurance athletes (Dempsey et al. 1984; Powers et al. 1984; Williams et al. 1986). Although it is well documented that EIH can occur in the healthy athlete, no data exist to describe the incidence of EIH in this population. Information concerning the incidence and under which exercise conditions that EIH occurs would be of obvious benefit to the physician or physiol- ogist performing exercise testing. Therefore, the purpose of the present investigation was to quan- tify the occurance of EIH in healthy adults with particular reference to the elite endurance ath- lete. Methods Subjects. Sixty-eight subjects were divided into three groups according to their fitness level: 1) untrained (UT), N= 16; 2) moderately trained (MT), N= 27; and 3) elite highly trained endurance athletes (EHT), N=25. The EHT subjects selected for study were chosen on the basis of outstanding running (N=12) or cycling (N=13) performances (Vo.... >68 ml 9 kg-1. min-1). All were healthy, non-smoking, adult males with no history of lung disease and normal pulmonary func- tion at the time of testing. The UT group was physically active but were not engaged in a regular exercise program. The MT group was composed of non-athletes who perform regular aerobic exercise 3 to 7 times per week. The EHT group had been performing endurance training for several years and were actively competing in distance running or cycling at the time of testing. Descriptive data for each group is contained in Ta-