Intraosseous versus intravenous vascular access during cardiopulmonary resuscitation for out-of-hospital cardiac arrest: a systematic review and meta-analysis of observational studies

Table 2 Results of the summary effect estimates for the primary outcome

Analysis	Outcome or moderator variables	Number of included studies	Odds ratio (95% confidence interval)	I², %	p for mixed-effects meta-regression analysis
Main analysis	Favorable neurological outcome at hospital discharge	6 [19, 20, 22,23,24,25]	0.60 (0.27–1.33)	95	NA
Subgroup analysis
Study period	Before 2010	1 [19]	0.22 (0.11–0.42)	0	0.19
Study period	After 2010	5 [20, 22,23,24,25]	0.73 (0.33–1.63)	94
Study region	North America	4 [19, 20, 23, 25]	0.60 (0.20–1.79)	94	> 0.99
Study region	Europe	2 [22, 24]	0.57 (9.94E-06-32,578.24)	94
Selection of vascular access	First attempted vascular access	3 [19, 20, 25]	0.49 (0.09–2.66)	94	0.56
Selection of vascular access	Actual vascular access	3 [22,23,24]	0.73 (0.07–7.22)	95
EMS response time	Longer than 10 mins	1 [22]	1.30 (0.97–1.73)	0	0.29
EMS response time	Shorter than 10 mins	5 [19, 20, 23,24,25]	0.51 (0.21–1.26)	94
Time to intervention	Timing adjusted in analysis	4 [20, 22, 23, 25]	0.89 (0.48–1.66)	89	0.02
Time to intervention	Timing not adjusted in analysis	2 [19, 24]	0.22 (0.17–0.30)	0
Type of effect estimates pooled in analysis	Adjusted effect estimates	5 [19, 20, 22, 23, 25]	0.71 (0.30–1.65)	95	0.25
Type of effect estimates pooled in analysis	Unadjusted effect estimates	1 [24]	0.23 (0.11–0.50)	0
Initial arrest rhythm	Shockable rhythms	4 [19, 20, 23, 25]	0.64 (0.27–1.53)	90	NA

ISSN: 1757-7241