The impact of excluding adverse neonatal outcomes on the creation of gestational weight gain charts among women from low- and middle-income countries with normal and overweight BMI.

BACKGROUND: Existing gestational weight gain (GWG) charts vary considerably in their choice of exclusion/inclusion criteria, and it is unclear to what extent these criteria create differences in the charts' percentile values. We aimed to establish the impact of including/excluding pregnancies with adverse neonatal outcomes when constructing GWG charts.

METHODS: This is an individual participant data analysis from 31 studies from low- and middle-income countries. We created a dataset that included all participants and a dataset restricted to those with no adverse neonatal outcomes: preterm < 37 weeks, small or large for gestational age - SGA or LGA, low birth weight < 2,500 g, or macrosomia > 4,000 g. Quantile regression models were used to create GWG curves from 9 to 40 weeks, stratified by pre-pregnancy BMI, in each dataset.

RESULTS: The dataset without the exclusion criteria applied included 14,685 individuals with normal weight and 4,831 with overweight. After removing adverse neonatal outcomes, 10,479 individuals with normal and 3,466 individuals with overweight remained. GWG distributions at 13, 27, and 40 weeks were virtually identical between the datasets with and without the exclusion criteria, except at 40 weeks for normal weight and 27 weeks for overweight. For the 10th and 90th percentiles, the differences between the estimated GWG were larger for overweight (approximately 1.5 kg) compared to normal weight (< 1 kg). Removal of adverse neonatal outcomes had minimal impact on GWG trajectories of normal weight. For overweight, the percentiles estimated in the dataset without the criteria were slightly higher than those in the dataset with the criteria applied. Nevertheless, differences were < 1 kg and virtually nonexistent at the end of pregnancy.

CONCLUSIONS: Removing pregnancies with adverse neonatal outcomes had little or no influence on the GWG trajectories of individuals with normal and overweight.