Embolic activity during in vivo cardiopulmonary bypass.

Neurologic injury after cardiac surgery is principally associated with emboli. Although much work has focused on surgical sources of emboli, less attention has been focused on emboli associated with the heart-lung machine. We tested whether emboli are associated with discrete processes during cardiopulmonary bypass (CPB). One hundred patients undergoing cardiothoracic surgery were enrolled between April 2008 and May 2011 at a single medical center. During each surgical procedure, emboli were counted in three CPB locations: the venous side (Channel 1), before the arterial line filter (Channel 2), and after the arterial line filter (Channel 3). We used prespecified event markers to identify perfusionist interventions. Identical circuits were used on all patients. Of the 100 patients enrolled, 62 underwent isolated coronary artery bypass grafting (CABG), 17 underwent isolated valve operations, and 21 underwent CABG plus valve. Median counts across Channels 1, 2, and 3 were 69,853, 3,017, and 1,251, respectively. The greatest contributor to emboli in Channels 1, 2, and 3, respectively, were achieving the calculated CPB flow, opening of the electronic arterial line clamp, and introducing a hemofilter. The circuit technology was efficient in reducing total emboli counts from Channels 1-2 irrespective of the size of the emboli. Nearly 71% of all emboli 30-100 microm in size were removed from the circuit between Channels 2 and 3. No significant association was found between emboli counts and S100B release. Emboli occur frequently during CPB and are predominantly associated with the initiation of bypass, operation of the electronic arterial line clamp, and the initiation of a hemofilter. Continued work to reduce the occurrence of emboli is warranted.