Background: The protective effect of collateral vessels in coronary artery disease (CAD) is well established. Little is known, however, about factors that influence collateral formation.
Methods: We studied the coronary angiograms of 200 consecutive patients with single-vessel coronary artery occlusion. Patients were excluded if obstructive stenoses were present in other vessels or if prior revascularization had been undertaken. Collateral circulation to the occluded artery was graded as 'poor' (no or incomplete filling) or 'rich' (complete filling). Patient characteristics, including mode of presentation, medications and CAD risk factors, were assessed.
Results: Positive univariate correlates of rich collaterals included increasing age [odds ratio (OR) 1.03, P = 0.016], 'statin' use (OR 2.50, P = 0.005), nitrate use (OR 1.96, P = 0.034), calcium-channel blocker (CCB) use (OR 4.07, P < 0.001), presentation with stable angina (OR 2.34, P = 0.006), longer time since diagnosis of CAD (OR 1.12, P = 0.002) and history of hyperlipidemia (OR 3.55, P < 0.001). Significantly poorer collateralization was observed in the setting of acute myocardial infarction (MI) (OR 0.23, P < 0.001), diabetes mellitus (OR 0.33, P = 0.003), impaired left ventricular function (OR 0.64, P = 0.015) and occlusion of the left anterior descending coronary artery (LAD) (OR 0.28, P < 0.001). On multivariate analysis, rich collateralization was associated with hyperlipidemia (P = 0.003) and CCB use (P = 0.028). Independent predictors of poor collaterals were presence of diabetes (P < 0.001), LAD occlusion (P = 0.001) and presentation with acute MI (P = 0.017).
Conclusion: Diabetes mellitus, occlusion of the LAD and presentation with acute MI are independently associated with poor distal vessel collateralization, whereas hyperlipidemia and use of CCBs are associated with rich collateralization. Factors determining coronary collateral formation may in turn influence outcomes after coronary artery occlusion.