Aim: To investigate whether the serum levels of interleukin‐1β, 6, 8, tumour necrosis factor‐α and the soluble receptor of IL‐2 are useful in the diagnosis of neonatal sepsis, and whether their diagnostic power is increased when in combination with classical markers such as C‐reactive protein and white blood cell count. Methods: Blood samples were collected at admission from 40 neonates with suspected infection. Patients were included in different groups according to the bacteriological and laboratory results: Group I consisted of 20 newborns with positive blood cultures and other biological tests suggestive of infection. Group II included 20 neonates with negative blood cultures and biological tests not suggestive of infection. The control group included 20 healthy neonates with no clinical or biological data of infection. Results: Mean values of C‐reactive protein were significantly higher in Group I. No differences were found between the groups for white blood cell count, with the exception of the presence of leucocytosis in Group II. Levels of interleukin‐1β, 6, 8, tumour necrosis factor‐α, soluble receptor of interleukin‐2, and C‐reactive protein were significantly higher in infected neonates than in the control groups. Detection sensitivity and specificity were 80 and 92% for C‐reactive protein, 60 and 87% for interleukin‐1β, 61 and 80% for interleukin‐6, 62 and 96% for interleukin‐8, 54 and 92% for tumour necrosis factor‐α and 63 and 94% for soluble receptor of interleukin‐2. The discriminant analysis showed that the best combination for sepsis diagnosis was C‐reactive protein + interleukin‐8 + soluble receptor of interleukin‐2, with a sensitivity of 85% and a specificity of 97.1%. Conclusion: Our study suggests that no individual test can on its own identify infected neonates, and that although the combination of C‐reactive protein, interleukin‐8 and the soluble receptor of interleukin‐2 exhibits a high specificity, its sensitivity is limited.

The peripheral repertoire of CD4 1 T lymphocytes contains autoreactive cells that remain tolerant through several mechanisms. However, nonspecific CD41 T cells can be activated in physiological conditions as in the course of an ongoing immune response, and their outcome is not yet fully understood. Here, we investigate the fate of human naive CD4 1 lymphocytes activated by dendritic cells (DCs) presenting endogenous self-peptides in comparison with lymphocytes involved in alloresponses. We generated memory cells (Tmem) from primary effectors activated with mature autologous DCs plus interleukin (IL)-2 (Tm auto ), simulating the circumstances of an active immune response, or allogeneic DCs (Tm allo ). Tmem were generated from effector cells that were rested in the absence of antigenic stimuli, with or without IL-7. Tmem were less activated than effectors (demonstrated by CD25 downregulation) particularly with IL-7, suggesting that this cytokine may favour the transition to quiescence. Tm auto and Tm allo showed an effector memory phenotype, and responded similarly to polyclonal and antigen-specific stimuli. Biochemically, IL-7-treated Tm allo were closely related to conventional memory lymphocytes based on Erk-1/2 activation, whereas Tm auto were more similar to effectors. Autologous effectors exhibited lower responses to IL-7 than allogeneic cells, which were reflected in their reduced proliferation and higher cell death. This was not related to IL-7 receptor expression but rather to signalling deficiencies, according to STAT5 activation These results suggest that ineffective responses to IL-7 could impair the transition to memory cells of naive CD41 T lymphocytes recognizing self-peptides in the setting of strong costimulation.

No abstract

No abstract