The evidence basis of diving and hyperbaric medicine - a synthesis of the high level clinical evidence with meta-analysis

Download files
Access & Terms of Use
open access
Copyright: Bennett, Michael Heywood
Introduction: Hyperbaric oxygen therapy (HBOT) is the administration of 100% oxygen at pressures greater than 1 atmosphere. One recurrent criticism that has been made of this field is that treatment is based on little or no good clinical evidence. Aims: The primary objective of this thesis is to make a useful response to that criticism. I planned to collate all the available randomised evidence in the fields of diving and hyperbaric medicine, supply a critical appraisal of each paper, and synthesise that evidence in a series of systematic reviews with meta-analysis. I also intended to use a cost analysis of hyperbaric practice in our own facility to inform formal cost-effectiveness analysis using the estimates of effect generated by the individual meta-analyses. Methods: A comprehensive search strategy was used to identify all clinical RCTs involving the administration of hyperbaric breathing mixtures. Each trial was appraised using the software developed by the Oxford Centre for Evidence Based Medicine. Each critical appraisal was loaded onto a searchable web site at Each diagnostic category identified was considered for inclusion in a Cochrane systematic review and meta-analysis. Results: The database includes 130 critical appraisals covering 173 separate reports. The site has received more than 17,000 hits. There are 12 formal meta-analytical reviews and all have been accepted for publication in the Cochrane Database of Systematic Reviews at the time of writing. These form the basis of this thesis and include late radiation tissue injury, chronic wounds, acute hearing loss and tinnitus, multiple sclerosis and decompression illness. The meta-analyses in this thesis suggest there are several areas where HBOT is associated with improved clinical outcomes and that routine use is probably justified in some areas (e.g. radiation proctitis healing with HBOT: NNT 3, 95%CI 2 to 11). On the other hand, these analyses suggest there is most unlikely to be significant clinical benefit from the application of HBOT to patients currently referred for HBOT (e.g. multiple sclerosis). Conclusions: The randomised evidence for the use of HBOT is now significantly easier to access. Recommendations for therapy and future research directions can be made on the basis of these analyses.
Persistent link to this record
Link to Publisher Version
Additional Link
Bennett, Michael Heywood
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
Resource Type
Degree Type
PhD Doctorate
UNSW Faculty
download whole.pdf 6.25 MB Adobe Portable Document Format
Related dataset(s)