BACKGROUND: Verify the accuracy of interleukin 6 (IL-6) and C-reactive protein (CRP) for diagnosis of late onset sepsis in newborn (NB) infants. METHODS: a prospective cohort study with 43 NB infants hospitalized at the NICU with suspicion of late onset sepsis was carried out. Levels of IL-6 and of CRP were dosed with suspicion diagnoses; day (0) and sequentially on day 1, 3, and 7 of the evolution and the best cut-off values were calculated for the diagnoses. Indices of sensibility (S), specificity (SP), positive predictive value and negative predictive value (PPV, NPV) were calculated for each test as well as for the combination between them. RESULTS: Levels of IL-6 and CRP were above the established cut-off values in the NB infants with sepsis and with presumed sepsis. There was a significant difference between both groups, where the only difference was the positive blood culture for the first group. Diagnosis could be rejected in 6 NB infants. The IL-6 showed better indices on the day of suspicion diagnosis, day 0 (S: 88.9%, Sp: 80.0%, PPV: 76.2%, NPV: 90.9%), followed by the C-reactive protein (S: 94.0%, Sp: 78.3%, PPV: 77.3%, NPV: 94.7%) 24 hours later. The combination of IL-6 / CRP demonstrated to be adequate for early diagnosis of sepsis on day 0 and 24 hours later with S and NPV of 100%. CONCLUSION: for diagnosis of sepsis the combination interleukin 6 / CRP presented accuracy. During the following days their development reflected the clinical evolution of the NB infants.
Newborn infant; Late onset sepsis; Diagnosis; Interleukin 6; C-reactive protein
