Both in-water recompression on oxygen and normobaric oxygen are associated with reduced bubble counts when administered at one hour after a provocative dive

1. Both IWR and NBO were associated with reduced bubble counts after an 82 minute exposure.

2. Clinical importance is unclear. The research question has not been answered.
Citation/s: Dituri J, Sadler R, Siddiqi F, Sadler C, Javeed N, Annis H, Whelan H. Echocardiographic evaluation of intracardiac venous gas emboli following in-water recompression. Undersea & hyperbaric medicine: journal of the Undersea and Hyperbaric Medical Society, Inc. 2016 Mar 1;43(2):103-12.

Lead author's name and fax: J. Dituri. Contact Harry Whelan:
Three-part Clinical Question: For divers suffering decompression sickness, is in-water recompression a useful alternative emergency treatment to normobaric oxygen administration when a recompression chamber is not available?

Search Terms: In-water recompression; decompression sickness; oxygen
The Study:

Single-blinded (outcome assessor) randomised controlled trial with intention-to-treat.

The Study Patients: Adult technical or commercial divers who were given intervention one hour after surfacing from a standard provocative dive to 33.5 metres for 25 minutes.

Control group (N = 5 ; 5 analysed): Cardiac echo at one hour after surfacing, then 100% oxygen breathing on the surface for 82 minutes.

Experimental group (N = 6 ; 6 analysed): As above but IWR breathing 100% oxygen at 9 metres then 7.5 metres for a total of 82 minutes.
The Evidence:

NBO Group
IWR Group
Bubbles per frame after completion of oxygen breathing period
0 – 3.8
0 – 0.2

1. Authors state the main outcome was the average reduction of echo bubble score in each dive group (presumably before and after treatment). They have failed to give this result in the paper. 2. None of the secondary outcomes (test subject observations and physical exam on a neurological checklist) were significantly different. No figures given.

3. There seems to be some conflation between 'decompression stress' and the treatment of actual decompression illness symptoms in the introduction.

4. One subject (group not specified) had suffered multiple episodes of DCI in the past, and three subjects had a PFO demonstrated.

5. No apparent attempt at sample size calculation. Study has very low power to detect important differences between groups.

6. No control group (untreated) so it is not clear if bubble counts would have been reduced with no intervention.
Appraised by: Mike Bennett; Wednesday, 22 November 2017

Kill or Update By: November 2020