Xu2017

Hyperbaric oxygen therapy improved the proportion good outcomes in diabetic patients following haemorrhagic stroke

1. More individuals who had HBOT were deemed to have a good outcome at six months following the haemorrhagic stroke 2. Glycaemic control was improved with HBOT at six months

Citation/s:1. Xu Q, Fan SB, Wan YL, Liu XL, Wang L. The potential long-term neurological improvement of early hyperbaric oxygen therapy on haemorrhagic stroke in the diabetics. Diabetes research and clinical practice. 2018 Apr 1;138:75-80.
Lead author's name and fax: xuqian6521@126.com

Three-part Clinical Question:For patients with diabetes and suffering a haemorrhagic stroke, does the application of hyperbaric oxygen improve outcome?
Search Terms: intracerebral haemorrhage; stroke; diabetes

The Study:Double-blinded concealed randomised controlled trial with intention-to-treat.
The Study Patients: Adult patients with a history of diabetes and presenting to the emergency department within 24 hours of a CT-confirmed haemorrhagic stroke. Excluded if severe (NIHSS >22) or died over the next six months.
Control group (N = 32; 32 analysed): All standard care plus air breathing in a chamber at 1.5 ATA for 60 minutes daily for 30 days.
Experimental group (N = 47; 47 analysed): As above but breathing 100% oxygen at 2.5 ATA.

The Evidence:

Outcome	Time to Outcome	Sham rate	HBO rate	Relative Risk Reduction	Absolute Risk Reduction	NNT
Modified Rankin Score -good outcome (< 2) Range 0 (best) to 5 (worst)	6 months	0.69	0.89	30%	0.21	5
	95% CIs:			3% to 57%	0.02 to 0.39	3 to 44
Good Outcome NIHSS (< 3) Range 0 (Best) to 42 (Worst)	6 months	0.56	0.81	44%	0.25	-4
	95% Confidence Intervals:			-80% to -7%	0.04 to 0.45	2 to 25
Barthel Index 95 to 100 Range 0 (worst) to 100 (Best)	6 months	0.66	0.85	-30%	-0.195	-5
	95% CIs:			0% to 59%	0.001 to 0.39	3 to 682

 

Non-Event Outcomes	Time to outcome/s	Control group	Experimental group	P-value
HbA1c	6 months	9.1 SD 0.82	8.0 SD 0.79	< 0.05

Comments:
1. Intriguing study with interesting outcomes. It should be reproduced elsewhere sooner rather than later as the results seem very important
2. Note planned retrospective exclusion of patients who died within 6 months, but none died.
3. Patients were randomised on a 1 Control :1.5 HBO schedule.
4. Modified Rankin <2 is no symptoms or no significant disability, NIHSS <3 similar and Barthel Index suggests fully independent functioning

Appraised by: Mike Bennett m.bennett@unsw.edu.au; Thursday, 23 January 2020
Kill or Update By: January 2023