Pathogenetic lessons from experimental models of necrotizing enterocolitis

Abstract

Wei Hseuh and her colleagues' work on platelet activating factor (PAF) and tumour necrosis factor (TNF) [3] highlights the complex interrelationship between these inflarnmatory mediators, oxygen-derived free radicals, arachidonic acid metabolites, and the complement system in producing bowel necrosis in animals. However, we feel that the evidence implicating these mediators particularly TNF, in clinical necrotizing enterocolitis (NEC) is less convincing than in these animal studies. The amounts of PAF and TNF injected in these studies correspond to plasma levels of 68 gg/ml and 113 gg/ml, respectively, assuming a blood volume of 80 ml/kg and a packed cell volume of 0.45, which is greatly in excess of the levels detected by them clinically. In contrast to Caplan, Harris [1] was unable to detect a significant rise in TNF at the time of diagnosis in infants with either bacterial sepsis or NEC. We have similarly found no significant difference in plasma TNF concentration in 14 infants with NEC (Bell stage 2 or 3) during the first 48 h (41+29 pg/ml) compared to controls (40 + 31 pg/ml, unpaired t = 0.07, df = 44, p -0.94) as measured by ELISA. Neither do our results support the suggestion that TNF is released as a result of inflammatory damage, as levels from 48 to 306 h were similarly low (42_ 32 pg/ml). If TNF is involved in the pathogenesis of NEC, it may be disappearing from the circulation at the time of diagnosis, when other cytokines such as interleukin 6 are rising [2].