Plasma albumin and haemodilution: the problem of interpretation in sequential studies

The ability of four formulae (3 based on packed cell volume and one on alpha-2 macroglobulin) to correct for postoperative dilutional changes in plasma albumin was assessed in 4 patients undergoing coronary artery surgery. Cardiopulmonary bypass is a stringent test of the efficacy of the formulae to correct for haemodilution as a large volume of fluid is added to the circulation in a relatively short period of time. Plasma albumin was measured 8 times in the perioperative period in the 4 patients and the clinical situation was mimicked in vitro by adding various volumes of diluent to whole blood. None of the formulae consistently corrected for predicted changes in plasma albumin concentration in vivo; two of the formulae corrected for in vitro changes in plasma albumin but the same formulae produced considerable over-correction when applied in vivo. The problem of correcting for perioperative dilutional changes in plasma albumin is discussed.