Two recent studies have reported an association between antenatal exposure to co-amoxyclav, either alone or in combination with erythromycin, and neonatal necrotising enterocolitis (NEC).^1,2 Based on the analyses of secondary outcomes in these studies, the authors raised concerns about the use of co-amoxyclav antenatally and recommended further investigation of its use in the neonatal period.

We have completed a case-control study designed to test the hypothesis that perinatal exposure to co-amoxyclav is associated with an increased risk of NEC. During a 17 year period (1983–2000), 32 cases of NEC were diagnosed in preterm infants born to mothers inbooked at Liverpool Women’s Hospital. Of these, 17 were diagnosed at laparotomy, 12 had classical radiological features, and in three the diagnosis was made on clinical grounds alone. Two gestation matched controls were selected for each index case. Information on potentially relevant perinatal variables, including antenatal and postnatal exposure to co-amoxyclav, were collected from maternal and infant case notes.

Infants who developed NEC tended to be lighter at birth (median birth weight 853 (interquartile range (IQR) 717–1248) g v 1037 (IQR 779–1613) g in controls, p = 0.065) and were more often delivered after absent or reversed flow identified on umbilical artery Doppler studies (p = 0.007). Postnatally, Gram negative septicaemia preceding NEC was significantly more common in cases than controls (p = 0.005). However, the frequency of perinatal exposure to co-amoxyclav was similar in both groups (table 1).

In summary, there is no evidence from this study of a link between perinatal exposure to co-amoxyclav and NEC. Our findings do not support the hypothesis that treatment with co-amoxyclav is causally associated with the development of NEC.