Utility of Sonoclot in Prediction of Postoperative Bleeding in Pediatric Patients Undergoing Cardiac Surgery for Congenital Cyanotic Heart Disease: A Prospective Observational Study
Abstract
Objectives
To assess the utility of Sonoclot in prediction of postoperative bleeding in pediatric patients undergoing cardiac surgery with cardiopulmonary bypass for congenital cyanotic heart disease.
Design
Prospective, observational study.
Setting
Single university hospital.
Participants
Eighty-seven pediatric patients undergoing cardiac surgery for congenital cyanotic heart disease.
Interventions
Laboratory coagulation parameters (prothrombin time, international normalization ratio, activated partial thromboplastin time, fibrinogen, D-dimer) as well as point-of-care Sonoclot glass bead activation time, clot rate, and platelet function (gbPF) were done before induction of anesthesia and following heparin reversal after termination of cardiopulmonary bypass (CPB) in all patients.
Measurements and Main Results
Postoperative blood loss was monitored by the amount of chest tube drainage. The primary outcome was to define Sonoclot parameters for prediction of postoperative bleeding. Secondary outcomes studied were amount of postoperative blood loss, transfusion requirement of various blood products, incidence of surgical re-exploration, duration of postoperative mechanical ventilation, intensive care unit and hospital stay. Among studied subjects, 37.9% (33 of 87 patients) were designated as bleeders while 62.1% (54 of 87 patients) were non-bleeders. Lower age, D-dimer, and gbPF test after termination of CPB following heparin neutralization were predictive for postoperative bleeders. Among these, post-protamine gbPF had the highest area under the curve (0.725), 95% confidence interval (0.619-0.831) for prediction of postoperative bleeders. Duration of mechanical ventilation (26.41 ± 36.44 v 8.25 ± 6.36 h, respectively, p = 0.001), intensive care unit stay (7.36 ± 4.05 v 4.96 ± 2.49, p = 0.001), and hospital stay (11.69 ± 4.82 v 8.63 ± 3.48 p = 0.001) were higher in bleeders; however, incidence of re-exploration was comparable between both groups.
Conclusion
Postoperative bleeders may be predicted independently by post-CPB gbPF, postoperative D-dimer, and lower age of patients. Among these, post-CPB gbPF has maximum predictive value.