A high neonatal serum eosinophil cationic protein level is a risk factor for atopic symptoms

To show whether neonatal eosinophil counts (EC) and eosinophil cationic protein (ECP) can be used in assessing the risk of atopy, alone or in combination with family history of atopy (FHa).

A group of 63 newborns was included: 38 with FHa, 25 without FHa. A blood sample was collected on the 4th day of life for EC and ECP evaluation. Clinical follow-up was conducted after 1, 3, 6, 12, 18, 24 and 36 mo. The chi2 test and the Student's t-test were used to compare dichotomic and continuous variables, respectively. Variables shown to be significant by univariate analysis were evaluated with a multivariate regression model and the relative risks (RR) were estimated.

Twenty-six newborns (41%) displayed atopic manifestations during the follow-up. Twenty-one (55%) of 38 newborns with FHa displayed atopic symptoms versus 5/25 (20%) without FHa (p 0.012). Neither EC nor serum ECP levels were significantly different between newborns with FHa and newborns without. ECP levels did not differ between newborns with single heredity and newborns with dual heredity. EC did not differ significantly between newborns who developed atopy and those who did not. Instead, serum ECP levels were significantly higher in newborns who developed atopy (mean 27.9 microU/l vs 16.8 microU/l). Atopic manifestations appeared in 16 (62%) of 26 newborns with ECP > or = 18 microU/l compared with 10 (27%) of 37 with ECP < 18 microU/l (p = 0.006). In the multivariate regression model, with ECP < 18 microU/l and no FHa as reference class, the class 1 (no FHa and ECP > or = 18 microU/l) has a low RR (1.4), class 2 (FHa and ECP < 18 microU/l) an intermediate RR (2.7) and class 3 (FHa and ECP > or = 18 microU/l) a very high RR (16.3).

Neonatal serum ECP levels, in contrast with EC, were significantly higher in newborns who developed atopic manifestations during follow-up. The risk of atopy was about twice as great when ECP was > or = 18 microU/l (and four times as great in multivariate analysis). When serum ECP was combined with FHa, the RR for newborns with FHa and ECP > or = 18 microU/l was 16 times greater than for those without FHa and ECP < 18 microU/l. The identification of newborns at "extremely high atopic risk" (FHa and ECP > or = 18 microU/l) may be expecially useful--in clinical practice--in newborns with only one atopic parent or sibling, for whom it is not universally agreed that dietary and environmental prevention measures should be applied.