Objective: Children with congenital heart diseases undergoing surgery with cardiopulmonary bypass are exposed to a high risk of perioperative endotoxemia. The aim of our study was to prospectively evaluate endotoxin assay activity reference levels during the postoperative phase of infants undergoing cardiac surgery for congenital heart disease and to assess their association with perioperative variables and postoperative infections.
Design: Prospective exploratory single-center cohort study.
Setting: Tertiary pediatric cardiac ICU.
Patients: Infants undergoing cardiac surgery with cardiopulmonary bypass were enrolled. Exclusion criteria were preoperative suspected or confirmed infection, the need for extracorporeal membrane oxygenation or a ventricular assist device in any perioperative phase, surgery for heart transplantation, and/or urgent surgery.
Interventions: Serial measurements of endotoxin assay activity were performed at baseline, pediatric cardiac ICU arrival, postoperative day 1 and 2.
Measurements and main results: Twenty-five patients were enrolled. Overall, 14 of 25 patients (58%) presented at least one endotoxin assay activity level greater than 0.4 during the study period (normal level is < 0.4). Endotoxin assay activity levels tended to significantly increase from baseline to postoperative day 1 and 2 and from pediatric cardiac ICU arrival to postoperative day 2 (p < 0.0001). Endotoxin assay activity greater than 0.6 predicted Gram-negative infections with a sensitivity of 0.40, a specificity of 0.95, a positive predictive value of 0.66, and a negative predictive value of 0.86. At multivariable regression, endotoxin assay activity on postoperative day 1 resulted independently associated with cardiopulmonary bypass duration, lactate, temperature peak, and vasoactive inotropic score at pediatric cardiac ICU arrival. Children with endotoxin assay activity levels greater than 0.6 (vs all the others) showed a significantly higher median (interquartile) number of ventilation days: 8 (2-39) versus 1.5 (1-3 (p = 0.02).
Conclusions: This exploratory study showed that endotoxin assay activity levels in infants undergoing cardiopulmonary bypass are frequently above 0.4 and peak 24-48 hours after surgery and appear to be associated with perioperative impaired organ perfusion. Endotoxin assay activity is not useful to predict Gram-negative infections.