The temporal trends and preprocedural predictors of emergency coronary artery bypass graft surgery (ECABG) after elective percutaneous coronary intervention (PCI) in the contemporary era are largely unknown. From January 2003 to December 2014 elective hospitalizations with PCI as the primary procedure were extracted from the Nationwide Inpatient Sample. ECABG was identified as CABG within 24 hours of elective PCI. Temporal trends of elective PCI, ECABG, comorbidities, and in-hospital mortality were analyzed. Logistic regression model was used to identify preprocedural independent predictors of ECABG and post-PCI ECABG risk score was developed using the regression coefficients from the logistic regression model in the development cohort. The score was then validated in the validation cohort. Of 1,605,641 elective PCI procedures included in the final analysis, 5,561 (0.3%) patients underwent ECABG. The incidence of ECABG, co-morbidities and overall in-hospital mortality increased over the study period, whereas the in-hospital mortality after ECABG remained unchanged. An increasing trend of elective PCI performed at facilities without on-site CABG was noted, with a higher unadjusted in-hospital mortality in this cohort. ECABG risk score, performed well with a significantly higher risk of ECABG in those patients with a score in the highest tertile compared with those with lower ECABG score (0.6% vs 0.3%, p = 0.0005). In conclusion, an increasing trend of adverse outcomes after elective PCI is observed. We describe an easy-to-use predictive score using preprocedural variables that may allow the operator to triage the patient to an appropriate setting in an effort to improve outcomes.
Copyright © 2020 Elsevier Inc. All rights reserved.