Association of atheroma as assessed by intraoperative transoesophageal echocardiography with long-term mortality in patients undergoing cardiac surgery

Aims To determine whether the presence and severity of aortic atheroma predict long-term all-cause mortality among patients undergoing cardiac surgery. Methods and results We followed 8581 patients who underwent cardiac surgery and had routine intraoperative transoesophageal echocardiography for 2.8 years (range 0.06-6.0 years). Data regarding multiple potential confounders were prospectively collected and electronically recorded. There were 2878 (34%) patients with no atheroma; 4129 (48%) patients with mild atheroma; 1215 (14%) with moderate atheroma; and 359 (4%) with severe atheroma. There were 1000 deaths. Death rates were increased in patients with moderate [relative risk (RR) 3.29, 95% CI 2.50-4.32, P < 0.0001) and severe atheroma (RR 5.21, 95% CI 3.65-7.41, P < 0.0001). After adjusting for multiple other confounders, severe atheroma remained modestly predictive of risk (adjusted RR 1.46, 95% CI 1.07-2.00, P = 0.02); but moderate atheroma and mild atheroma were not predictive of increased risk. In a propensity analysis that matched patients with comparable range of variables, severe atheroma was no longer predictive of risk (adjusted RR 1.39, 95% CI 0.87-2.23, P = 0.17). Conclusion Our study shows that severe atheroma is associated with increased long-term mortality in patients undergoing cardiac surgery; however, the relationship is weak using propensity analysis, suggesting no causal association.