Lin2

The use of hyperbaric oxygen therapy in sub-acute patients with head injury was not associated with significant improvements in either Glasgow Outcome or Glasgow Coma Scores.

Clinical Bottom Line:
1. No significant improvement in either Glasgow Outcome Score or Glasgow Coma Scale following treatment.

Citation/s:1. Lin JW, Tsai JT, Lee LM, Lin CM, Hung CC, Hung KS, Chen WY, Wei L, Ko CP, Su YK, Chiu WT. Effect of hyperbaric oxygen on patients with traumatic brain injury. Acta Neurochrirugica 2008; 101(Supplement):145-149 . Lead author's name and fax: Wen-Ta Chiu [| wtchiu@tmu.edu.tw]

Three-part Clinical Question: For patients with sub-acute head injury, does the addition of hyperbaric oxygen to standard rehabilitation therapy result in improved functional outcomes? Search Terms: head injuries, closed, Glasgow outcome score

The Study:Non-blinded randomised controlled trial without intention-to-treat. The Study Patients: Adult patients with a history of moderate/severe head injury and not requiring mechanical ventilation. No significant chest or abdominal injuries. Control group (N = 31; 22 analysed): An unspecified standard approach to care and rehabilitation. Experimental group (N = 31; 22 analysed): As above plus 100% oxygen at 2ATA for 90 minutes daily for twenty days over four weeks.

The Evidence:

Outcome Time to Outcome Control rate Hyperbaric rate Relative risk reduction Absolute risk reduction Numbers needed to treat
Glasgow outcome score improved 3 months 0.23 0.29 28% 0.06 16
^ 95% CI: 124% to -68% 0.28 to - 0.15 NNT 4 to INF NNH 7 to INF
Glasgow outcome score improved 6 months 0.29 0.39 33% -0.097 10
^ 95% CI: 114% to -47% 0.33 to - 0.14 NNT 3 to INF NNH 7 to INF
Measure Control Group Hyperbaric Group Difference 95% CI
^ Mean SD Mean SD ^ ^
Glasgow coma score after therapy 11.5 ? 13.5 ? -2.0 N/A

Comments: 1. Unusual design and it is not clear how randomisation was achieved. Each randomised patient was assigned a control matched for sex, age and severity. 2. Authors report significant improvements in both GCS after therapy and GOS at six months, but we cannot confirm these differences with the information given. 3. Patients who did not tolerate HBOT were excluded from analysis, along with their matched controls. 4. No standard deviations, 95% confidence intervals or power calculations are given. 5. In general, this study has a number of methodological and reporting problems.

Appraised by: Mike Bennett, Sydney ; Tuesday, 27 January 2009. Email: [| m.bennett@unsw.edu.au] Kill or Update By: November 2021

Sumhorsa.gif

BACK