Background Preterm infants tend to be lighter and shorter on reaching term equivalent age (TEA) than those born full-term, but the effect of preterm birth on body composition (BC) is less clear. We conducted a systematic review to assess the effect of preterm birth on BC at TEA.

Methods The databases MEDLINE, EMBASE, CAB Abstracts, HMIC, ‘Web of Science’ and ‘CSA-Conference Papers Index’ were searched between 1947 and January 2011, with selective citation and reference searching. Included studies needed to have directly compared the BC of preterm infants (born <37 weeks) at TEA with infants born full-term. Meta-analysis was performed using the random effects model.

Results Seven studies were included. There were 371 infants in the preterm group and 291 in the full-term group. Gestational ages and weights at birth (mean±SD) were 30.0±2.7 weeks and 1.23±0.31 kg in the pre-term group and 39.3±1.3 weeks and 3.45±0.61 kg in the full-term group.

Meta-analysis showed that preterm infants had significantly lower weight, length and head circumference at TEA (mean differences 570 g, 3.8 cm and 1 cm respectively, p<0.00001 for all). The preterm infants had greater total body fat per unit body weight at TEA than those born full-term (mean difference 2.7%, p=0.05).

Conclusions Preterm infants are shorter and lighter at TEA, with a significantly greater proportion of body fat relative to body weight compared to infants born full-term. These findings demonstrate that weight is an insufficient criterion for assessing nutritional outcome in preterm infants, and robust measures of BC are required in addition to standard anthropometry.