Matteo Paganini ^1* Lorenzo Zucchi ¹ Tommaso Antonio Giacon ¹ Luca Martani ¹ Simona Mrakic-Sposta ² Giacomo Garetto ³ J. Chris McKnight ⁴ Enrico M. Camporesi ⁵ Richard E. Moon ⁶ Gerardo Bosco ^1*

• ¹ Department of Biomedical Sciences, University of Padova, Padova, Veneto, Italy
• ² Institute of Clinical Physiology, Department of Biomedical Sciences, National Research Council (CNR), Pisa, Tuscany, Italy
• ³ ATIP Centro di OssigenoTerapia Iperbarica, Padova, Italy
• ⁴ Scottish Oceans Institute, School of Biology, Faculty of Science, University of St Andrews, St. Andrews, Scotland, United Kingdom
• ⁵ Team Health Research Institute, Tampa, Florida, United States
• ⁶ Duke Center for Hyperbaric Medicine and Environmental Physiology, Department of Anesthesiology, School of Medicine, Duke University, Durham, North Carolina, United States

Background: Current diving physiology postulates that SCUBA divers’ arterial blood gas (ABG) levels vary proportionally to environmental pressure, but, to date, ABGs have only been obtained during simulated dives. Also, recent evidence supports the use of the arterial/alveolar (a:A) partial pressures of oxygen (PO2) ratio to predict the arterial PO2 (PaO2) under hyperbaric conditions from measurements obtained at 1 atmosphere absolute (ATA). This work summarizes ABGs obtained in SCUBA divers in real underwater conditions and aims to validate the a:A ratio in predicting PaO2 in this subset of individuals at depth. Methods: The study was approved by the local ethics committee. After cannulating the radial artery of the non-dominant limb, ABGs were sampled at the surface before the dive (A), at depth (15 meters of freshwater (mfw) or 42 mfw) before (B) and after (C) pedaling on a submersed bicycle for 10 minutes, and back at surface (D). After calculating the surface alveolar PO2 for each subject, the a:A ratio was obtained and used to predict PaO2 at depth. A linear regression between measured and predicted PaO2 was reported, along with the goodness-of-fit F test. Results: Six subjects performed the dive at 15 mfw, and four others at 42 mfw. The PaO2 proportionally increased at both depths, remaining stable before and after pedaling. The a:A calculated from the baseline ABG obtained at rest, out of the water, adequately predicted the PaO2 at depth (R2 = 0.97 , p<0.001), better at 15 mfw but losing accuracy at 42 mfw. Conclusions: The ABGs confirmed the proportional rise of PaO2 in SCUBA divers underwater. The a:A ratio could be used to predict the magnitude of PaO2 rise at depth to limit exposure to hyperoxia, especially in repetitive recreational dives and professional divers.