Removal of glucose from the cardiopulmonary bypass prime: a prospective clinical audit

To quantify our decision for the removal of glucose and the use of mannitol as a substitute osmotic agent in the cardiopulmonary bypass prime, we conducted a prospective clinical audit to evaluate the effects of this change on patient outcomes. Data were prospectively collected for 172 consecutive routine cardiac surgery patients. The first 85 patients (Surgeon A, 42 patients [Group 1], Surgeon B, 43 patients [Group 2]) received 1000 mL Plasmalyte 148 + 5% glucose as per institutional protocol. The remaining priming volume for each group consisted of 500 mL hemaccel or 4% albumin, 50 mL 8.4% sodium bicarbonate, 100 mL Hartmann's solution. The change to a glucose-free prime was then initiated, substituting Plasmalyte 148 (without 5% glucose) for the Plasmalyte 148 + 5% glucose, in addition 12.5 g mannitol was administered following delivery of cardioplegia to the patients operated on by Surgeon B. Surgeon A would not include mannitol at this time. Forty-one patients operated by Surgeon A (Group 3) subsequently received Plasmalyte 148, and 46 patients operated on by Surgeon B (Group 4) received Plasmalyte 148 plus mannitol. Analysis was performed stratified by surgeon to quantify the effects of removing glucose from the prime. Comparisons were made between groups 1 and 3, and 2 and 4. Net fluid changes were recorded from pre-CPB, up to 24-h postoperatively. Intraoperative data collection included serum glucose, hematocrit, osmolality, return to rhythm, arrhythmias, and blood transfusions. Post-operative variables, including cardiac enzymes, arrhythmias, intubation time, length of stay, and mortality were also collected. Removal of glucose from the CPB prime resulted in a lower serum glucose concentration (mmol/L) during CPB (Gp 1 [13.6] vs. Gp 3 [5.4]; Gp 2 [14.7] vs. Gp 4 [5.4], p < .05). The addition of 12.5 g of mannitol to the CPB prime resulted in a significantly lower net fluid gain (mL) 24 h postoperatively (Gp 2[2792] vs. Gp 4 [1970], p < .05) and greater CPB hematocrit (%) (Gp 2 [24.3] vs. Gp 4 [26], p < .05). No other results were found to be significant (except CPB plasma osmolality (Groups 2 and 4) and sodium concentration [Groups 1 and 3]). The results of our audit provide an evidence base to support our change in practice to utilize nonglucose primes.