Resumen

BENJUMEA, María Victoria. Diagnostic accuracy of five gestational references to predict insufficient birth weight. Biomédica [online]. 2007, vol.27, n.1, pp.42-55. ISSN 0120-4157.

Introduction. The maternal anthropometry is a potentially valuable tool in the evaluation of pregnancy status and prediction of birth weight. Objectives. Diagnostic accuracy of birth weight was compared for five gestational anthropometric references. Materials and methods. Longitudinal study was designed for 245 pregnant Colombian participants. Variables consisted of birth weight, gestational age, maternal height and weight. The references were as follows: gestational weight gain of Fescina (Uruguayan samples), % weight/height and body mass index of Rosso-Mardones (Chilean sample), pregestational weight of the Institute of Medicine (United States sample), and the body mass index of Atalah and colleagues (Chilean sample). The insufficient birth weight (<3,000 g) was defined as the critical threshold value. The following indices, with correspondent confidence intervals (CI 95%) were obtained: diagnostic accuracy, sensitivity, specificity, positive and negative predictive values, and agreement (Kappa index).   Results. The ability to predict the weight <3,000 g was higher in the first trimester measurement with the Chilean references, and in the second trimester with the USA and the Uruguayan references. The best concordance was presented between the Chilean references and the least with the Uruguayan. After the second trimester, the Chilean reference of Atalah and colleagues provided the best diagnostic accuracy, specificity and positive predictive value. Conclusion. The gestational anthropometric references that provided the most heterogeneous results were the US Institute of Medicine and the Fescina. Overall, the Atalah reference was the most accurate diagnostic predictor of insufficient weight at birth.

Palabras clave : Anthropometry; infant very low birth weight; body mass index; sensitivity and specificity.

        · resumen en Español     · texto en Español     · Español ( pdf )