Hyperbaric oxygen therapy reduced pain associated with interstitial cystitis

1. Mean pain score reduced following HBOT
2. Trend toward benefit from HBOT for urgency, frequency, bladder volume and subjective response, but no statistically significant differences

Citation/s:
1. van Ophoven A, Rossbach G, Pajonk F, Hertle L. Safety and efficacy of hyperbaric oxygen therapy for the treatment of interstitial cystitis a randomised, sham controlled, double blind trial. Journal of Urology 2006; 176:1442-1446.
Lead author's name and fax: avo@uni-muenster.de

Three-part Clinical Question: For patients with interstitial cystitis, does the application of hyperbaric oxygen result in any improvement in symptoms?
Search Terms: Cysitis, interstitial, pelvic pain

The Study:
Double-blinded randomised controlled trial with intention-to-treat.
The Study Patients: Adult women with a diagnosis of interstitial cystitis.
Control group (N = 7; 7 analysed): Baseline therapy for each individual was continued plus air breathing at 1.3 to 1.4 ATA for 90 minutes daily to 30 sessions over 5 weeks.
Experimental group (N = 14; 14 analysed): As above but 100% oxygen breathing at 2.4 ATA.

The Evidence:

Outcome

Time to Outcome

Sham rate

HBO rate

Relative risk reduction

Absolute risk reduction

NNT

Symptomatic responders

3 months

0

0.21

INF

0.21

5

95% Confidence Intervals:

 

-0.001 to 0.43

NNT = 2 to INF

 NNH = 1,194 to INF

 

Measure

Sham Group

HBO Group

Difference

95% CI

Mean

SD

Mean

SD

Interstitial cystitis index at 3 months

26.7

5.8

19.9

7.9

6.8

-0.28 to 13.88

Pain score (VAS) 3 months (0-100)

55.7

27.8

31.2

19.8

24.5

2.57 to 46.43

Bladder volume 3 months (mls)

118

36

147

49

29

-14.60 to 72.60

Comments:
1. The difference in the proportion of symptomatic responders on a subjective scale (3 HBO versus 0 sham) is reported in the paper as statistically significant (P<0.05), but our calculations do not support this finding (P=0.55 by Fisher's exact test, double one-sided P method). {Authors agree with the published figure is incorrect.}
2. Underpowered study with no sample size calculation.
 

Appraised by: Sean Hopson and Mike Bennett, Sydney ; Friday, 10 November 2006
Email: m.bennett@unsw.edu.au
Kill or Update By: March 2007