Rc. Silveira et Rs. Procianoy, Evaluation of interleukin-6, tumour necrosis factor-alpha and interleukin-1 beta for early diagnosis of neonatal sepsis, ACT PAEDIAT, 88(6), 1999, pp. 647-650
The objective of this study was to assess the contribution of interleukin-6
(IL-6), tumour necrosis factor-alpha (TNF-alpha) and interIeukin-1 beta (I
L-1 beta) to an early diagnosis of early-onset neonatal sepsis. A cohort of
117 newborn infants delivered during a 1-y period had IL-6, TNF-alpha and
IL-1 beta, blood and cerebrospinal fluid (CSF) cultures, leucocyte and plat
elet count collected on the initial evaluation of possible early-onset seps
is. They were divided into four groups: I, positive blood and/or CSF cultur
es; II, probably infected with clinical sepsis but negative cultures; III,
same as group II but mother received antibiotic antepartum; and IV, newborn
infants that did not receive any antibiotic therapy. There were no differe
nces among the four groups with respect to mean gestational ages and birthw
eights, median Apgar scores, type of delivery, or number of newborn infants
with leucocyte count <5000 mm(-3) or >25 000 mm(-3), platelet count <100 0
00 mm(-3), immature/total neutrophil ratio >0.2, absolute neutrophil count
<1000 mm(-3) and median IL-1 beta levels. Median IL-6 and TNF-alpha levels
were significantly higher in groups with patients with a diagnosis of clini
cal sepsis than in controls. The optimal cut-off point was 32 pg ml(-1) for
IL-6 and 12 pg ml(-1) for TNF-alpha. The combination of both provided a se
nsitivity of 98.5%. In conclusion, the combination of IL-6 and TNF-alpha is
a highly sensitive marker of sepsis in the immediate postnatal period.