Incidence and predictors of postoperative ischaemic stroke after coronary artery bypass grafting in the United States

Abstract

Abstract Background Postoperative acute ischemic stroke (AIS) is a catastrophic complication of coronary artery bypass grafting (CABG). There is limited data on the incidence and outcomes of AIS complicating CABG in the contemporary era, and whether these have changed over the years. Purpose To study the incidence and outcomes of postoperative AIS in a nationally representative cohort of CABG procedures over a 12-year period and examine predictors of AIS in patients undergoing CABG. Methods The National Inpatient Sample was used to identify all adult patients (>18 years old) who underwent CABG in the United States between January 2004 and September 2015. Multivariable logistic regression was performed to examine the associations between postoperative AIS and in-hospital mortality and identify predictors of AIS after CABG, expressed as odds ratios (OR) with corresponding 95% confidence intervals (CI). Results A total of 2,569,597 CABG operations were analysed. The incidence of postoperative AIS was 1.8% (n=47,279) in the overall cohort increasing from 1.2% in 2004 to 2.3% in 2015 (p<0.001). Significantly higher rates of AIS were observed amongst patients with atrial fibrillation (AF) and those undergoing non-elective or concomitant valve operations over the study period (see Figure). Patient risk profiles increased over time in both AIS and no-AIS cohorts, with higher Charlson comorbidity scores observed amongst AIS patients. AIS was independently associated with increased odds of in-hospital mortality (OR 3.03, 95% CI 2.93, 3.13) and prolonged hospital stay (∼6 more days) and a higher hospitalisation cost (∼$80,000 more) (p<0.001 for all). Several factors were predictors of AIS including age>60 years (61–70 years: OR 1.33, 95% CI 1.29, 1.37; 71–80 years: OR 1.49, 95% CI 1.44, 1.54; >80 years: OR 1.42, 95% CI 1.37, 1.48), female sex (OR 1.33, 95% CI 1.31, 1.36) and AF (OR 1.14 95% CI 1.12, 1.16) (p<0.001 for all). In contrast, on-pump CABG was not an independent predictor of stroke (OR 1.01, 95% CI 0.94, 1.09) (p=0.784). Conclusion In this nationally representative study, we have shown that the rates of postoperative stroke following CABG have increased over time in line with complexity of patient risk profiles. The present findings emphasise the need for further work on strategies to reduce the risk of postoperative stroke after CABG. Trends of postoperative AIS (2004-2015) Funding Acknowledgement Type of funding source: None