Twenty-seven RCTs (n=2,979) were included.
Four studies scored 5 points on the Jadad scale, two scored 4, five scored 3, and the rest scored either 2 or 1. Three studies used triple-blinding, four used double-blinding, and the remainder used single-blinding.
There was no evidence of statistically significant heterogeneity for mortality, MI or myocardial ischaemia (p>0.10). Significant heterogeneity was found for hospital and ICU length of stay, duration of mechanical ventilation, cardiac index, troponin level and post-bypass inotrope administration.
There was no statistically significant difference between patients receiving volatile and non-volatile anaesthetic agents for MI (12 RCTs), mortality (5 RCTs), myocardial ischaemia (8 RCTs) or length of ICU stay (10 RCTs).
Patients receiving volatile anaesthetics compared with non-volatile anaesthetics had significantly higher cardiac indices (random-effects model, WMD 0.22, 95% CI: 0.06, 0.38, p<0.006); significantly lower troponin I serum concentrations (random-effects model, WMD -1.44, 95% CI: -2.34, -0.55, p<0.002); significantly lower requirements for inotropic support (random-effects model, OR 0.50, 95% CI: 0.31, 0.80, p<0.004); significantly shorter duration of mechanical ventilation (random-effects model, WMD -2.71 hours, 95% CI: -5.30, -0.12, p<0.04); and a significantly shorter hospital stay (random-effects model, WMD -1.05 days, 95% CI: -1.68, -0.43, p<0.001).