Balanced ultrafiltration: inflammatory mediator removal capacity

Ultrafiltration with a hemoconcentrator may remove excess fluid load and alleviate tissue edema and has been universally adopted in extracorporeal circulation protocols during pediatric cardiac surgery. Balanced ultrafiltration is advocated to remove inflammatory mediators generated during surgery. However, whether balanced ultrafiltration can remove all or a portion of the inflammatory mediator load remains unclear. The inflammatory mediator removal capacity of zero-balanced ultrafiltration was measured during pediatric extracorporeal circulation in vitro. Extracorporeal circulation was composed of cardiotomy reservoir, D902 Lilliput 2 membrane oxygenator, and Capiox AF02 pediatric arterial line filter. The Hemoconcentrator BC 20 plus was placed between arterial purge line and oxygenator venous reservoir. Fresh donor human whole blood was added into the circuit and mixed with Ringer's solution to obtain a final hematocrit of 24-28%. After 2 h of extracorporeal circulation, zero-balanced ultrafiltration was initiated and arterial line pressure was maintained at approximately 100 mmHg with Hoffman clamp. The rate of ultrafiltration (12 mL/min) was controlled by ultrafiltrate outlet pressure. Identical volume of plasmaslyte A was dripped into the circuit to maintain stable hematocrit during the 45 min of the experiment. Plasma and ultrafiltrate samples were drawn every 5 min, and concentrations of inflammatory mediators including interleukin-1β (IL-1β), IL-6, IL-10, neutrophil elastase (NE), and tumor necrosis factor-α (TNF-α) were measured. All assayed inflammatory mediators were detected in the ultrafiltrate, demonstrating that the ultrafiltrator may remove inflammatory mediators. However, dynamic observations suggested that the concentration of NE was highest among the five inflammatory mediators in both plasma and ultrafiltrate (P < 0.001). IL-1β had the lowest concentration in plasma, whereas the concentration of TNF-α was the lowest in ultrafiltrate (P < 0.001). Concentrations of all inflammatory mediators in the ultrafiltrate did not increase linearly compared with those in plasma. The respective ultrafiltrate to plasma concentration and amount ratios indicated that the total removal effect of hemoconcentrator on the inflammatory mediators was 4.17 ± 2.68% for IL-1β, 0.64 ± 0.69% for IL-6, 0.24 ± 0.18% for IL-10, 2.84 ± 1.65% for NE, and 0.51 ± 0.81% for TNF-α, respectively. Balanced ultrafiltration may selectively remove inflammatory mediators from serum. Respective ratios of inflammatory mediators in ultrafiltrate compared with plasma, as well as total amount of inflammatory mediators in the ultrafiltrate suggest that balanced ultrafiltration removes a limited portion of the total inflammatory mediator load.