Adverse gastrointestinal complications after cardiopulmonary bypass: can outcome be predicted from preoperative risk factors?

UNLABELLED: Adverse gastrointestinal (GI) outcome after cardiac surgery is an infrequent event but is a clinically important health care problem because of associated increased morbidity and mortality. The ability to identify patients at greatest risk before surgery may be helpful in planning appropriate perioperative management strategies. We examined the pre- and intraoperative characteristics of 2417 patients from 24 diverse United States medical centers enrolled in the Multicenter Study of Perioperative Ischemia Study who were undergoing cardiac surgery using cardiopulmonary bypass as predictors for adverse GI outcome. Resource utilization was evaluated for patients with and without adverse GI outcomes. Adverse GI outcomes occurred in 5.5% of patients (133 of 2417), increased in-hospital mortality 6.5-fold, prolonged the mean intensive care unit length of stay by 1 wk, and more than doubled the mean postoperative hospital stay (P < 0.0001). Predictors of adverse GI outcome included decreased left ventricular function, hyperbilirubinemia, thrombocytopenia, prolonged partial thromboplastin time, prior cardiovascular surgery, combined coronary artery bypass graft surgery and intracardiac or proximal aortic surgery, pharmacological cardiovascular support, and intraoperative transfusion. The literature suggests that adverse GI outcome after cardiac surgery is secondary to poor splanchnic perfusion, which many of these risk factors may predict. Therefore, patients deemed to be at risk before surgery may benefit from tightly controlled hemodynamic management and other strategies that optimize perioperative organ perfusion. IMPLICATIONS: We identified the preoperative and intraoperative predictors associated with an increased incidence of postoperative gastrointestinal complications after cardiac surgery using cardiopulmonary bypass. Because these complications are associated with frequent morbidity and mortality, these predictors may be helpful in identifying patients at increased risk so that risk stratification can be modified and perioperative management can be appropriately adjusted.