Hypoxia Signatures in Closed-Circuit Rebreather Divers

Daniel Popa, UCSD, Hamilton Award Recipient 2018

Malfunctions in closed circuit rebreathers (CCRs) can cause hypoxia if oxygen is not added to the breathing loop and the diver remains unaware of decreasing oxygen levels. Hypoxia, dangerously low oxygen levels, can lead to confusion, loss of consciousness underwater and drowning. Pilots in a low-pressure (hypobaric) exposure have shown a “hypoxia signature” with the same constellation of hypoxia symptoms each time they are exposed to hypobaric hypoxia. This study examines CCR divers’ ability to recognize their hypoxia signature and perform self-rescue.


Closed circuit rebreather (CCR) diving has proliferated recently but carries a significantly higher mortality risk than open-circuit scuba diving. Hypoxia is one of the leading causes of rebreather diving injuries and fatalities, particularly since symptoms come on gradually as the diver consumes oxygen in the breathing circuit.

Among aviators, acute hypobaric hypoxia can present with a variety of symptoms with substantial differences between individuals. Interestingly, each pilot’s hypobaric hypoxia symptoms with repeated exposures appears reproducible and serves as a “hypoxia signature.”

This study investigates these hypoxic signatures in an equal number of CCR and scuba divers. They are asked to identify a hypoxic trial while blinded to the gas mix they are breathing from a CCR after a non-blinded trial of hypoxia. Each subject undergoes an experimental day of four trials with the first trial non-blinded and stimulating hypoxia. We prevent the CCR from adding oxygen to the breathing loop so the subject consumes the oxygen in the loop and gradually becomes hypoxic, mimicking a real world CCR failure. After this episode of known hypoxia, each subject undergoes three additional randomized, blinded trials with one hypoxic trial and two sham trials (i.e., the CCR functions correctly and maintains room air). While breathing from a CCR, each subject pedals a cycle ergometer at low wattage to simulate swimming underwater. While the subject breathes and cycles, they perform a validated computer-based neurocognitive test, “Go/No-Go,” to track decision-making capacity, reaction time and cognitive inhibition. This test further mimics underwater distractions. In addition, each subject points to common hypoxia symptoms listed on a board as they appear in real time during each trial to record each subject’s hypoxic signature. During each trial, we monitor the subject’s pulse oximetry, breath-by-breath end tidal CO2 and O2, and inspired O2 from the CCR. If at any point the subject’s O2 saturation drops to 75%, we terminate the trial.

If at any time the subject feels that they are experiencing a dive emergency from which they would normally perform a self-rescue procedure, the subject turns a ball valve to simulate switching to a bail-out gas mixture to demonstrate the cognitive ability to perform self-rescue. Upon completion of each trial, we ask the subject if they thought the blinded trial was a sham or hypoxia and why. After completion of each of the four trials, we debrief further in order to educate the subject on which of the blinded trials was hypoxic versus sham.

Twenty of a proposed 30 subjects have undergone the study protocol. 95% of these subjects showed agreement between unblinded and blinded hypoxia symptoms, and 85% of these subjects correctly identified the gas mixture of the trial.

During unblinded hypoxia, only 25% of these subjects performed unprompted bailout. Also, only 55% correctly performed a bail-out but only when prompted and 15% were unable to perform a bail-out procedure, even when prompted.

During blinded hypoxia, only 45% of subjects performed the bailout unprompted while 15% remained unable to bailout despite prompting.

In conclusion, even if these data support a normobaric hypoxia signature among both CCR and SCUBA divers, most subject were unable to recognize hypoxia and performed a self-rescue unprompted so these results do not support hypoxia exposure training for CCR divers.


This study was partially funded through the DAN/R.W. Hamilton Dive Medicine Research Grantin 2018. 

Additional Reading References

Scientific article

  • Popa, Daniel, Craig Kutz, UCSD Department of Emergency Medicine, Division of Hyperbaric and Undersea Medicine, San Diego, CA, USA, Morgan Carlile, UCSD Department of Emergency Medicine, Division of Hyperbaric and Undersea Medicine, San Diego, CA, USA, Kaighley Brett, Canadian Armed Forces, Toronto, Canada, et al. “Hypoxia Signatures in Closed-Circuit Rebreather Divers.” Diving and Hyperbaric Medicine Journal 52, no. 4 (December 20, 2022): 237–44. https://doi.org/10.28920/dhm52.4.237-244.

Hypoxia signatures in closed-circuit rebreather divers – PubMed (nih.gov)

Related publications

  • Fock, Andrew W. “Analysis of Recreational Closed-Circuit Rebreather Deaths 1998-2010.” Diving and Hyperbaric Medicine 43, no. 2 (June 2013): 78–85.
  • Cable, Gordon G. “In-Flight Hypoxia Incidents in Military Aircraft: Causes and Implications for Training.” Aviation, Space, and Environmental Medicine 74, no. 2 (February 2003): 169–72.
  • Mitchell, Simon J., Hayden M. Green, Stacey A. Reading, and Nicholas Gant. “The Utility and Safety of Hypoxia Experiences for Rebreather Divers.” Diving and Hyperbaric Medicine 49, no. 2 (June 30, 2019): 112–18. https://doi.org/10.28920/dhm49.2.112-118.

To view the publication, click here