Reduction of false-positive diagnosis of fetal growth restriction by application of customized fetal growth standards.

OBJECTIVE: To evaluate the clinical performance of fetal growth charts adjusted for individual maternal characteristics.
METHODS: The study group consisted of 267 low-risk singleton pregnancies with normal clinical outcome. Mothers were recruited prospectively after the booking visit, then underwent three to five ultrasound examinations for fetal weight estimation. Individual growth curves were generated from these data and the birth weight, based on logpolynomial growth model. Computer software was written to calculate the number of fetal growth curves that cross the tenth percentile limit, based on an unadjusted, average ultrasound standard for our population, compared with the number that cross this limit if it is customized for known pregnancy characteristics such as maternal height, booking weight, parity, and ethnic group.
RESULTS: Individual growth trajectories of this group of pregnancies with normal outcome were significantly less likely to cross below the tenth percentile for fetal weight when using customized growth charts than when the unadjusted standard was used (McNemar's test, P < .001).
CONCLUSION: The relationship between maternal characteristics and fetal size needs to be considered in the assessment of fetal growth. The use of a customized standard reduces the false-positive rate for the diagnosis of growth restriction in a normal population.