Morbidity and Mortality After Acute Myocardial Infarction After Elective Major Noncardiac Surgery

To develop parsimonious models of in-hospital mortality and morbidity risk after perioperative acute myocardial infarction (AMI).

Retrospective data analysis.

National Inpatient Sample (2008-2013), a 20% sample of all non-federal in-patient hospitalizations in the United States.

Patients 45 years or older who experienced perioperative AMI during elective admission for noncardiac surgery.

The study used a mixed principal components analysis and multivariate logistic regression to identify risk factors for in-hospital mortality after perioperative AMI. A model incorporating only preoperative risk factors, defined by the Revised Cardiac Risk Index (RCRI), was compared with a "full risk factor" model, incorporating a large set of preoperative AMI risk factors. The risk of post-AMI disposition to an intermediate care or skilled nursing facility, a marker of functional impairment, then was evaluated.

In the present study, 15,574 cases of AMI after elective noncardiac surgery were identified (0.42%, corresponding with 78,122 cases nationally), with a 12.4% in-hospital mortality rate. The "RCRI-only" model was the best-fit model of post-AMI in-hospital mortality risk, without loss of predictive accuracy compared with the "full risk factor" model (area under the receiver operator characteristic curve 0.80, 95% confidence interval [CI] [0.77-0.82] v area under the receiver operator characteristic curve 0.81, 95% CI [0.77-0.83], respectively). Post-AMI mortality risk was the highest for perioperative complications, including sepsis (odds ratio 4.95, 95% CI [4.32-5.67]). Conversely, functional impairment was best predicted by the "full-risk factor" model and depended strongly on chronic preoperative comorbidities.

The RCRI provides a simple but adequate model of preoperative risk factors for in-hospital mortality after perioperative AMI.