Around 80% of intrauterine growth restricted (IUGR) infants are born at term . When pregnancy is complicated by IUGR, there is, whether term or preterm, a clear association with an increase in neonatal mortality and neonatal morbidity (short and long term) [2–4]. The long term morbidity ranges from behavioral problems and minor developmental delay to spastic cerebral palsy [5–10]. However, not all studies, especially after excluding congenital anomalies, confirm these findings . Besides fetal asphyxia, meconium aspiration, fetal heart rate abnormalities and low Apgar score, also more admittances to and longer stays at neonatal intensive care units are reported. This might partly be related to a higher prevalence of hypoglycaemia, neonatal sepsis, hypothermia and haematological problems as thrombocytopenia and polycythemia in these neonates [12–14].
When a fetus is small for gestational age (SGA), defined on the basis of a birth weight below the 10th centile, there is the concern that the fetus might be afflicted by IUGR . As SGA is defined on the basis of an arbitrary chosen cutoff birth weight centile, not all infants falling below the 10th centile are abnormally small because of growth restriction. Many neonates with a birth weight below the 10th centile are representing the normal spectrum of fetal growth . Variation in birth weight is related to many factors as maternal height, weight, parity and fetal gender, but also ethnicity . For that reason optimal growth for any fetus should be related to the fetus' own individual optimal growth curve [17–19]. Intrauterine growth restriction has to be defined on further knowledge such as Doppler abnormalities as seen in placental perfusion, eventually in combination with abnormalities in cerebral perfusion [20, 21] and possibly also by neonatal measurements as the Ponderal Index [22, 23].
A reduction of fetal growth is exponentially associated with a higher perinatal mortality  and morbidity [25, 26]. Doppler umbilical artery studies have shown that absence of end diastolic velocities, indicative of IUGR based on severe placental insufficiency is associated with a higher rate of caesarean deliveries and an increased incidence of perinatal and neonatal mortality [27–30]. However, a normal umbilical artery Doppler study at term gestation might be falsely reassuring, while a normal cerebral artery study might identify the fetus not likely having a major adverse outcome .
Most of the growth restricted children experience an accelerated growth, especially of the head circumference, during the first 6 months after birth . However, this upward centile crossing or 'catch up growth' is not complete, even at the age of seven years . Moreover head circumference seems to correlate with cognitive outcome .
Long-term neurological and cognitive development of the IUGR infant at term have been studied extensively. The Ponderal Index among IUGR infants, but also among infants with a normal birth weight, is an independent predictor of neonatal morbidity: the lower the Ponderal Index the higher morbidity . Learning difficulties, defects in speech and mild neurological deficits and behavioral problems have been reported to occur more in term neonates born SGA [35, 36]. At school ages (7–8 years) temperamental differences and differences in play behavior are apparent , most probably contributing to increased rate of school failure found in IUGR infants.
Long-term morbidity might be resulting from subtle nutritional insults to the brain in utero. Although the brain growth spurt, being the most vulnerable period of the human brain, spans a broad period between mid pregnancy and 6 months of postnatal age [38, 39], it is shown that growth failure occurring around term shows a strong association with cognitive disturbances as a poorer mental and psychomotor development at two years of age . However, not all studies, even at preschool age show this trend of increased problems in growth restricted infants [41, 42]. Besides neurodevelopmental consequences it is now also clear that children who were undernourished during pregnancy (e.g. born with a birth weight more than 2 SD below the mean birth weight) and especially in combination having had a compensatory growth trajectory during childhood have an increased risk in later life for diabetes, hypertension and cardiovascular diseases .
Given the data from studies concerning the effect of under-nutrition on the brain and the effects on long-term cognitive and behavioral outcome, evaluation of the possible clinical benefit of early induction of delivery, pre-empting a detrimental effect of chronic under nutrition on the fetal brain intervention, is important. By such an intervention it might be possible to start earlier with a more optimal feeding, compensating for the poor intra-uterine environment. Induction of labor is very often common practice in cases of suspected IUGR [44, 45]. In the Netherlands at 33 up to 36 weeks of gestation, 63% of IUGR pregnancies were induced, whereas from 37 weeks onwards this percentage is 23%; more than double the percentage in non-IUGR pregnancies. In a Dutch obstetric cohort of 14.294 primigravid women with IUGR pregnancies, 29% of these pregnancies were induced . In these pregnancies complicated by IUGR, induction of labour was associated with an increased risk of instrumental deliveries and emergency caesarean section, but no difference in neonatal outcome immediately after birth was found.
At present, there is no uniformity on the management of women with IUGR at term. Although there is no doubt that the intra-uterine growth retardated fetus should be considered as high risk, and should be monitored, there is no consensus on which diagnostic methods to evaluate fetal condition and subsequent intervention is best. It is unclear whether in this situation either induction of labour or expectant management is beneficial for the mother and her baby, since evidence on the subject is lacking.
For preterm pregnancies complicated by intra-uterine growth retardation, an international randomised clinical trial recently showed that expectant management had little benefit over early delivery with respect to short term neonatal outcome . However, results of this trial cannot be extrapolated to the situation at term.
The lack of consensus on the subject in the Netherlands is demonstrated by the fact that in 2002 in women with a SGA child, labour was induced in 32% of these women, whereas labour started spontaneously in 56% of these women, the remaining 11% had an elective caesarean section. These data are based on actual birth weight, and the clinical situation is even more complicated by the fact that the antenatal diagnosis of a SGA child is often difficult to make and easily missed in clinical practice.
In view of this clinical dilemma, we propose a randomised clinical trial in which induction of labour is compared with expectant monitoring in women with a suspected IUGR child at term. We will compare maternal outcome, neonatal outcome and maternal quality of life, as well as costs. Moreover, we will collect, in both randomisation arms, data of the diagnostic tests used in fetal surveillance, i.e. fetal heart rate pattern, sonographic measurement of the amniotic fluid index and Doppler measurement of the umbilical artery and the fetal medial cerebral artery in women.