The search for objective criteria at the limit of viability

Defining the limit of viability among very preterm infants remains a difficult issue. A recent study suggested severity of intrauterine growth retardation as a decisive criterion, reporting very poor survival for infants born at

To check whether the above criterion can be confirmed in a similar analysis of German data.

Analyses were based on German population-based data from Lower Saxony, including preterm neonates (22-28 weeks) without severe congenital malformations (n = 3,453), born 1991-1999. For the definition of growth retardation postnatal (<3rd, <10th centile) and antenatal (<3rd centile) reference values were used.

Depending on the definition used, proportions of growth retarded infants varied considerably (4-16%). Survival rates increased from 0 to 64%, arguing against considering infants <28 weeks' GA as a homogenous group with regard to their odds for survival.

There is an association between severity of growth retardation and VLBW mortality in general, but overinterpreting results from single studies may lead to dangerous conclusions. Data from preterm infants should be stratified by week of gestation.