Many clinical and experimental data strongly support the role of immune mechanisms in the pathogenesis of childhood epilepsy. Following Pechadre's first observations with intramuscular immune globulin (IMIG), intravenous immune globulin (IVIG) has been employed in some forms of intractable childhood epilepsy (ICE), mainly in West syndrome (WS) and Lennox Gastaut syndrome (LGS), with good results. So far, 373 children suffering from ICE have been treated in 29 studies and 174 have responded favourably. Although these studies are heterogeneous and controls are lacking, most authors report similar responsiveness ranging from 30% to 50%. Several mechanisms have been suggested to account for the efficacy of IVIG in ICE including antiviral effect, substitutive therapy in patients with concomitant humoral immunodeficiency, idiotype-anti-idiotype interaction or a neuromodulant effect. To better define the real efficacy of IVIG in ICE in paediatric patients, a randomized, multicenter, double-blind clinical trial was started in 1993, including only patients suffering from WS and LGS. To date, only one double-blind trial had been carried out (with both adult and paediatric patients); it showed a clear trend in favour of IVIG treatment but lacked statistical significance, perhaps because of the small and heterogeneous sample. Controlled multicentre studies on well-defined populations are needed and patients with WS and LGS are probably the best candidates.