Fetal size vs growth: comparative analysis of 3 models of growth velocity based on third trimester estimated fetal weights for identifying stillbirth risk

Abstract

Fetal growth velocity is being recognised as an important parameter by which to monitor fetal wellbeing, in addition to assessment of fetal size. However there are different models and standards in use by which velocity is being assessed. We wanted to investigate three clinically applied methods of assessing growth velocity and their ability to identify stillbirth risk, in addition to that associated with small for gestational age. Retrospective analysis of prospectively recorded, routine-care data of pregnancies with 2 or more third trimester scans in New Zealand. Results of the last two scans were used for the analysis. The models investigated to define slow growth were A. 50+ centile drop between measurements, B. 30+ centile drop and C. estimated fetal weight below a projected optimal weight range, based on pre-defined, scan interval specific cut-offs to define normal growth. Each method's ability to identify stillbirth risk was assessed against that associated with small-for-gestational age at last scan. The study cohort consisted of 71,576 pregnancies. The last 2 scans in each pregnancy were performed at an average of 32+1 and 35+6 weeks gestation. The three models defined 'slow growth' at differing rates: A.: 50 centile drop 0.9%; B.: 30 centile drop 5.1%; C.: below projected optimal weight range 10.8%. Neither of the centile based models identified at-risk cases that were not also small for gestational age at last scan. The projected weight range method identified an additional 79% of non-SGA cases as slow growth, and these were associated with a significantly increased stillbirth risk (RR: 2.0; 95% CI: 1.2-3.4). Centile based methods fail to reflect adequacy of fetal weight gain at the extremes of the distribution. Guidelines endorsing such models might hinder the potential benefits of antenatal assessment of fetal growth velocity. A new, measurement-interval specific projection model of expected fetal weight gain can identify fetuses that are not small for gestational age, yet at risk of stillbirth because of slow growth. The velocity between scans can be calculated using a freely available growth rate calculator (www.perinatal.org.uk/growthrate).