Inflammatory mediator removal by zero-balance ultrafiltration during cardiopulmonary bypass

The abnormal conditions to which blood is subjected during cardiopulmonary bypass (CPB) trigger an activation of the inflammatory response in all patients to varying degrees. Both complement activation and the release of cytokines characterize this response. Most inflammatory mediators have a molecular weight that is below the membrane pore size of commonly used ultrafilters, which should allow them to be freely filtered. However, some mediators have been shown to fail to cross through the membrane even though they are small enough to cross. The purpose of the present study was to determine whether certain inflammatory mediators could be removed by ultrafiltration when performed during the rewarming phase of CPB. Thirty adult patients undergoing a single, open-heart procedure were randomized to either control (no ultrafiltration) or to the zero-balance ultrafiltration (ZBUF) group. ZBUF was performed by removing 3 l/m2 blood using a 65-kDa ultrafilter with 1.3-m2 surface area. A volume of a balanced salt crystalloid solution (Plasmalyte) equal to the filtered blood volume was given to replace the fluid removed. Patient data was taken before CPB (T1), immediately following CPB (T2), and 12 h following the procedure (T3). The average volume of filtrate removed during ZBUF was 6405 ml, which was analyzed for the presence of interleukin (IL)-1, IL-6, tumor necrosis factor-alpha (TNF-alpha), C3a, and C5a. The average concentrations of the mediators measured in the effluent were: IL-1, 0.17 pg/ml; IL-6, 0.64 pg/ml; TNF-alpha, 1.25 ng/ml; C3a, 782.6 ng/ml; C5a, 25.6 ng/ml. In every case except for IL-1, the amounts of mediators removed were significantly greater than zero. This study demonstrates that ultrafiltration is a strategy that can be used during CPB in the adult to remove significant amounts of inflammatory mediators.