Improving cardiopulmonary bypass: does continuous blood gas monitoring have a role to play?

The CDI 500 (Terumo Cardiovascular Systems, Ann Arbor, MI) is an in-line blood gas monitoring device that has been used in clinical practice for over a decade. Few randomized studies have evaluated the value of this device with respect to improved perfusion management. We routinely use automated continuous quality indicator programs to assess perfusion management. The aim of this study is to investigate in a prospective randomized trial the role of in-line blood gas monitoring in the improvement of blood gas management during cardiopulmonary bypass (CPB) utilizing continuous quality indicators. Patients were randomized into two groups (Control, CDI). Patients in the Control group received our standard CPB blood gas management, with intermittent blood gas results. Continuous blood gas measurements from the CDI 500 were recorded at 20-second intervals, with the perfusionist blinded to these measurements. Patients in the CDI group received standard CPB blood gas management, in addition to continuous blood gas measurements visible on the CDI 500, the alarm system activated, and the data recorded. Perfusion management for all cases was guided by institutional protocols. One hundred patients (50 in each group) were included in the study. No significant difference existed between the groups on demographic, surgical, or clinical outcomes. Blood gas levels of patients in the CDI group were able to be maintained in accordance to protocol a greater percentage of the time, e.g., pCO2 management was 2% versus 20% (p = .008); this was most notable for differences between the Control and the CDI group for pCO2 > 45 mmHg (p = .003). Practice variation determined via statistical control charts improved for both pH and pCO2, represented by a decrease in the variation associated with practice. Continuous blood gas monitoring with the CDI 500 results in significantly improved blood gas management as determined by adherence to institutional protocols.