The association of perioperative transfusion with 30-day morbidity and mortality in patients undergoing major vascular surgery

Abstract

Objective: Blood transfusions are common among patients undergoing major vascular surgery. Prior studies suggest an association between blood transfusion and increased morbidity and mortality among patients undergoing cardiac surgery. The predictors of perioperative transfusion and its impact on patients undergoing vascular surgery have been poorly defined.

Methods: We examined data from a large multicenter quality improvement vascular surgical registry of all patients undergoing elective or urgent open peripheral arterial disease procedures, endovascular aneurysm repair, or open abdominal aortic aneurysm (AAA) repair between January 2012 and December 2013. Emergency cases, carotid endarterectomy, and carotid artery stenting were excluded. Univariate and multivariate logistic regression modeling was used to identify predictors of transfusion and association of transfusion with outcomes. All regression models had Hosmer-Lemeshow P > .05 and area under the receiver operating characteristic curve of >0.8, confirming excellent goodness of fit and discrimination.

Results: Our study population comprised 2946 patients who underwent open peripheral arterial disease procedures (n = 1744), open AAA repair (n = 175), or endovascular aneurysm repair (n = 1027) at 22 hospitals. The overall transfusion rate was 25%, at a median nadir hemoglobin level of 7.7 g/dL. Independent factors predicting transfusion included female gender (odds ratio [OR], 2.6; 95% confidence interval [CI], 2.1-3.2), nonwhite race (OR, 2.7; 95% CI, 1.4-5.2), preoperative admission status (ie, acute care hospital) (OR, 2.6; 95% CI, 1.3-5.3), preoperative anemia (OR, 4.2; 95% CI, 3.3-5.1), congestive heart failure (OR, 1.4; 95% CI, 1.1-1.9), prior myocardial infarction (OR, 1.3; 95% CI, 1.01-1.6), clopidogrel (OR, 1.4; 95% CI, 1.2-1.8), open AAA repair (OR, 25; 95% CI, 17-39), open bypass (OR, 3.5; 95% CI, 2.7-4.6), and urgent procedures (OR, 1.4; 95% CI, 1.1-1.8). With adjustment for major covariates, perioperative transfusion was independently associated with death (OR, 6.9; 95% CI, 3.2-15), myocardial infarction (OR, 8; 95% CI, 3.7-17), and pneumonia (OR, 7.4; 95% CI, 3.3-17).

Conclusions: Perioperative transfusion in vascular surgical patients is independently associated with increased 30-day morbidity and mortality. Given indeterminate causation, these data suggest the need for a prospective transfusion threshold study in vascular surgical patients.