Effect of Erythropoietin Treatment and Acetazolamide on Oxygen Saturation during Sleep at High Altitude

INTRODUCTION: Previous research has shown that Acetazolamide (AZ) increases oxygenation during sleep at high altitude. However, the effect of recombinant human erythropoietin (rEPO) and AZ on blood oxygen levels during sleep at high altitude has not been investigated. PURPOSE To determine the effect of rEPO+AZ compared to AZ only (Placebo; PL+AZ) on oxygen saturation (SaO 2 ), heart rate (HR), oxygen desaturation index (ODI), and Stanford Sleepiness scale score (SS). We hypothesize that there will be no difference of rEPO+AZ compared to Placebo+AZ on SaO 2 , HR, ODI, and SS during sleep at high altitude. METHODS Wrist oximeters were worn by 19 participants for one night at sea level baseline (SL; Natick, MA; 50m) and on high altitude (HA) nights 1 (HA1), 2 (HA2), 7 (HA7), and 14 (HA14) (Pike’s Peak, CO; 4300 m; 460 mmHg). After baseline measurements were collected, participants received rEPO (Procrit; n=9; 2 F) or PL (saline; n=10; 2 F) for 4-weeks (50 IU/kg BW; 3x/week). Before ascending to HA, all volunteers took AZ for 3 ½ days (250 mg bid) with the last full dose taken on HA. Participant sleepiness scores were reported using the SS each morning after wearing the wrist oximeters. Thirteen participants (rEPO+AZ n=6, PL+AZ n=7) provided SS responses. ODI was defined as a drop in SaO 2 by at least 4 percentage units for a minimum duration of 10 seconds. Statistical analyses were performed using linear mixed model to determine the main effects of treatment (rEPO+AZ vs. PL+AZ), time, and treatment x time interactions with baseline values as a covariate. RESULTS SaO 2 was 2 percentage units higher in the PL+AZ group compared to EPO+AZ (main effect of treatment, p = 0.002). SaO 2 was not different between HA1, HA2, and HA7, but increased by 2 percentage units from HA7 to HA14 (main effect of time; p = 0.035). There was no significant treatment x time interaction for ODI (p = 0.915). ODI was not different between the two groups (main effect of treatment, p = 0.368). ODI was not different from HA1 to HA2 nor from HA7 to HA14 but increased by 57 events per hour (main effect of time, p < 0.001) from HA2 and HA7. However, ODI was unchanged from HA7 and HA14 (p = 1.000). There was no difference in HR between groups (main effect of treatment, p = 0.557), over time (main effect of time p > 0.05 for all), or treatment x time interaction (p = 0.977). Similarly, for SS there was no main effect of treatment (p = 0.830), time (p = 0.530), or significant group x time interaction (p = 0.250). CONCLUSION Our primary novel finding is that SaO 2 was higher in the PL+AZ group compared to the rEPO+AZ group. Despite this difference in SaO 2 , participants’ SS were not different. More research is needed to investigate the potential mechanism for the higher oxygenation in the PL+AZ vs rEPO+AZ. Funded by USMRDC; authors’ views not representative of the US Army or DoD policy. This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.