In the promoter region of the IL-6 gene there is a single base exchange (G --> C) polymorphism at position -174. Recent findings suggest that this polymorphism may affect the transcription rate of the IL-6 gene and IL-6 plasma levels. To analyse its biological significance, we examined IL-6 plasma levels in cord blood and IL-6 production by neonatal cells after LPS-stimulation in relation to the presence of the IL-6G and IL-6C alleles. We hypothesized that since healthy neonates lack a previous exposure to exogenous antigens, their cytokine production could be genetically regulated. We also assumed that the normal labour-related stress could provide a physiological stimulus for IL-6 production. Cord blood was collected from 50 healthy, full-term neonates after normal vaginal delivery (VD) and from 42 healthy, full-term neonates after elective caesarean section (ECS). Adult samples were obtained from 450 healthy adult controls. The -174 polymorphism was analysed using PCR. IL-6 plasma levels and in vitro IL-6 production were measured using an ELISA method. Generally, IL-6 plasma levels in neonates were significantly higher than those in adults (neonates born by VD versus adults p < 0.001 and neonates born by ECS versus adults p < 0.001); the median value for neonates born by VD was 11.4 pg/ml (4.5-45.9), for neonates born by ECS 2.9 pg/ml (1.9-6.4) and for adults, 1.2 pg/ml (0.7-2.0). Surprisingly, cord blood IL-6 levels after VD differed significantly from those after ECS (p < 0.001). An analysis was carried out to ascertain if there was a genetic association between different IL-6 genotypes and IL-6 plasma levels in neonates. In the group of VD neonates with the CC genotype, non-carriers of the G allele, secreted significantly more IL-6 than carriers of the G allele (p < 0.03); 21.4 pg/ml (9.5-81.3) and 9.6 pg/ml (3.5-36.2) respectively. In line with this, ECS newborns with the CC genotype had higher IL-6 plasma levels than carriers of the G allele (p < 0.02); respective median values were 6.3 pg/ml (2.2-12.9) and 2.7 pg/ml (1.7-4.1). These findings were also supported when in vitro IL-6 production by neonatal mononuclear cells was compared carriers of the G allele and non-carriers of the G allele. IL-6 levels were significantly lower in carriers of the G allele than in non-carriers (p < 0.04); respective median values were 6,980 pg/ml (4,175-16,800) and 17,425 pg/ml (11,400-33,900). In vivo or in vitro production of IL-6 of adult controls was not associated with the IL-6 -174 polymorphism. The difference between cord blood IL-6 levels after VD and after ECS suggests that normal labour-related stress induces IL-6 production. Our data also suggest that the -174 polymorphism of the IL-6 gene participates in the regulation of IL-6 responses in both groups of neonates. Furthermore, the naive IL-6 response of stimulated neonatal cells is associated with the -174 polymorphism of the IL-6 gene. In healthy adults, the regulation of IL-6 responses differs from that of healthy neonates, since baseline and inducible IL-6 levels in adults were not associated with this polymorphism. This indicates that the genetic regulation of IL-6 production can be observed in naive cells, while in adult cells previous contact with exogenous antigens probably modifies their responses.