Modified ultrafiltration (MUF) is often used in conjunction with paediatric cardiac surgery with cardiopulmonary bypass (CPB) and is thought to improve clinical outcome. It is unclear whether these improvements (if any) are due to the removal of inflammatory mediators. In this prospective study, 18 children aged 12-24 months undergoing uncomplicated cardiac surgery with methylprednisolone added in the pump prime were randomized to receive CPB with (n = 10) and without (n = 8) MUF. Cytokines (TNFalpha, IL-6, IL-1beta, IL-10, IL-1ra), complement split products (C3d, C4d) and coagulation system activation (F1 + 2, ATIII) were measured pre-, peri- and up to 48 h postoperatively. For clinical outcome, the alveolar-arterial oxygen (A-a) gradient, transfusion requirement, drain loss, mean blood pressure and requirement for inotropic support were registered up to 24 h postoperatively. Our results show an improvement in postoperative oxygenation as well as a tendency towards decreased drain loss and improved haemodynamics in the MUF group. There were no intergroup differences detectable for TNFalpha, IL-1beta, IL-1ra, complement and coagulation markers. We conclude that MUF in itself does not significantly influence TNFalpha, IL-1beta, IL-1ra and the complement and coagulation profiles in children undergoing cardiac surgery with CPB. Despite this, there was some evidence for improved clinical outcome. Our results do not support that MUF improves postoperative organ function by modulation of the measured markers of inflammation.