Two RCTs (n=611 total, range 101 and 510) were included in the review.
Preoperative coronary revascularisation was associated with significantly higher rates of 30-day all-cause mortality (OR 2.01, 95% CI 1.04 to 3.89), 30-day death/myocardial infarction (OR 1.84, 95% CI 1.21 to 2.80) and late death/myocardial infarction (OR 1.60, 95% CI 1.16 to 2.21) than optimal medical therapy. There was no statistically significant difference between the groups for 30-day non-fatal myocardial infarction.
Compared to medical therapy, PCI was associated with a significantly higher rate at 30 days of non-fatal myocardial infarction (OR 2.14, 95% CI 1.05 to 4.36, I2=42.8%, random-effects model) and death/myocardial infarction (OR 2.44, 95% CI 1.54 to 3.82). There were non-significant trends to inferior results for other outcomes.
Compared to medical therapy, CABG was associated with non-significant trends to worse results for 30-day outcomes and a trend to superior results for late death/myocardial infarction.
When CABG and PCI were compared, CABG was significantly superior for late death/myocardial infarction (OR 0.60, 95% CI 0.37 to 0.98). There was a non-significant trend that favoured CABG for 30-day death/myocardial infarction.
Heterogeneity was low (I2=0% to 9.7%) for all subgroup analyses except those mentioned above.