M. Hayhoe R. Bellomo G. Liu L. McNicol B. Buxton The aetiology and pathogenesis of cardiopulmonary bypass-associated metabolic acidosis using polygeline pump prime Received: 21 October 1998 Final revision received: 16 March 1999 Accepted: 26 March 1999 Funding support: Austin and Repatriation Medical Centre, Department of Anaesthesia and Intensive Care Research Fund M. Hayhoe ´ R. Bellomo ( ) ) Intensive Care Unit, Department of Intensive Care Medicine, Austin and Repatriation Medical Centre, Heidelberg Vic 3084, Australia Tel.: + 61 (3) 9496±5992 Fax: + 61 (3) 9496±3932 email: rb@austin.unimelb.edu.au G. Liu ´ L. McNicol Department of Anaesthesia, Austin and Repatriation Medical Centre, Heidelberg Vic 3084, Australia B. Buxton Department of Cardiac Surgery, Austin and Repatriation Medical Centre, Heidelberg Vic 3084, Australia Abstract Objective: The pathogen- esis of the metabolic acidosis of car- diopulmonary bypass (CPB) is not fully understood. New quantitative methods of acid-base balance now make it possible to describe it more clearly. Accordingly, we studied acid-base changes during CPB with polygeline pump prime and defined and quantified the factors which contribute to metabolic acidosis. Design: Prospective cohort study. Setting: Tertiary institution. Participants: 10 cardiac bypass graft surgery patients. Interventions: Sampling of arterial blood at four time intervals: post-in- duction, on CPB during cooling and rewarming, and at skin closure. Mea- surement of serum Na + ,K + , Mg ++ , Ca ++ , Cl ± , bicarbonate, and phos- phate concentrations, arterial blood gases, and serum albumin, lactate, and pyruvate concentrations at each collection point. Analysis of findings according to quantitative physico- chemical principles, including calcu- lation of the strong ion difference ap- parent, the strong ion difference ef- fective, and the strong ion gap (SIG). Measurements and main results: All patients developed a mild metabolic acidosis. The median serum stan- dard bicarbonate concentration de- creased from 25.0 mEq/l post-induc- tion to 22.3 mEq/l at cooling and 22.2 mEq/l at rewarming (p < 0.05). The standard base excess decreased from a median of 1.55 mEq/l prior to CPB, to ±2.50 mEq/l at cooling, ±1.65 mEq/l at rewarming and, ±0.85 mEq/l at skin closure (p < 0.001). This mild metabolic aci- dosis occurred despite a decrease in the median serum lactate concen- tration from 3.20 mEq/l post-induc- tion to 1.83, 1.80, and 1.58 mEq/l at the three other time points. The in- crease in the median serum chloride concentration from 104.9 mEq/l post induction to 111.0, 111.1, and 110.0 mEq/l at the subsequent time points (p < 0.0001) was the main cause of the acidosis. There was also a significant increase in the SIG of 3.8 mEq/l at cooling and rewarming (p < 0.0001), suggesting a role for other unmeasured anions (polyge- line) in the genesis of this acidosis. Conclusions: Using quantitative biophysical methods, it can be dem- onstrated that, in patients receiving a pump prime rich in chloride and polygeline, the metabolic acidosis of CPB is mostly due to iatrogenic in- creases in serum chloride concen- tration and unmeasured strong an- ions (SIG). Its development is par- tially attenuated by iatrogenic hyp- oalbuminaemia. Changes in lactate concentrations did not play a role in the development of metabolic aci- dosis in our patients. Key words Metabolic acidosis ´ Cardiopulmonary bypass ´ Polygeline ´ Chloride ´ Cardiac surgery ´ Lactate Intensive Care Med (1999) 25: 680±685 Ó Springer-Verlag 1999 ORIGINAL