Albumin versus crystalloid for pump priming in cardiac surgery: meta-analysis of controlled trials.

OBJECTIVES To determine the effects of pump priming fluid choice on platelets, fluid balance, and clinical outcomes. DESIGN Meta-analysis of controlled clinical trials. Primary endpoints were platelet counts, colloid oncotic pressure, on-bypass fluid balance, postoperative weight gain, and colloid usage. SETTING Cardiac surgery with cardiopulmonary bypass. PATIENTS Adult and pediatric patients undergoing cardiac surgery, including coronary artery bypass grafting, valve procedures, and correction of congenital cardiac anomalies. INTERVENTIONS Extracorporeal circuit priming with either albumin or crystalloid. MEASUREMENTS AND RESULTS The meta-analysis included 21 controlled trials with 1,346 total patients. Albumin prime significantly reduced the on-bypass drop in platelet counts. The pooled weighted mean difference in platelet count drop with albumin versus crystalloid prime was -23.8 x 10(9)/L (confidence interval [CI], -42.8 to -4.7 x 10(9)/L). The colloid oncotic pressure decline was also smaller when albumin rather than crystalloid was used for priming, with a pooled weighted mean difference of -3.6 mm Hg (CI, -4.8 to -2.3 mmHg) during bypass and -2.0 mmHg (CI, -2.9 to -1.1 mmHg) after surgery. Albumin prime correspondingly reduced on-bypass positive fluid balance (-584 mL; CI, -819 to -348 mL) and postoperative weight gain (-1.0 kg; CI, -0.6 to -1.3 kg) compared with crystalloid. Postoperative colloid usage was lower with albumin than crystalloid prime (-612 mL; CI, -983 to -241 mL). CONCLUSIONS Albumin prime better preserves platelet counts than crystalloid. Albumin also favorably influences colloid oncotic pressure, on-bypass positive fluid balance, postoperative weight gain, and colloid usage. The clinical significance of these observations merits further investigation.