Evaluation of fetal medicine foundation algorithm in predicting small-for-gestational-age neonates

Abstract

Objectives

To examine the performance of the Fetal Medicine Foundation (FMF) 2012 predictive model and of isolated biophysical markers (uterine artery pulsatility index and mean arterial pressure) for small-for-gestational-age (SGA), in patients from Rio de Janeiro, Brazil.

Methods

For this cross-sectional study, SGA was diagnosed when a newborn presented birth weight below the fifth percentile for gestational age. FMF2012 algorithm sensitivity and specificity, positive (PPV) and negative (NPV) predictive value, positive likelihood ratio (LR +) and area under the ROC curve (AUC) were calculated to predict total and preterm SGA (SGA < 37). The performance of isolated biophysical markers – mean arterial pressure (MAP) and mean uterine artery pulsatility index (UtAPI) were studied.

Results

The final sample consisted of 1480 cases: 69 (4.6%) developed SGA, including 12 patients (0.8%) who were SGA < 37. The AUC showed that the performances of the FMF2012 combined model for SGA prediction was 0.687 and for preterm SGA was 0.824. With risk cutoff of 1:150, SGA screening yielded the following: sensitivity, 47%; specificity, 75%; LR +, 1.88; PPV, 8.66%; NPV, 96.72%. When screening for preterm SGA, we found sensitivity 66.6%, specificity 74.59%, LR +: 2.58, PPV 2%, and NPV 99.63%.

Conclusions

Performance of the FMF2012 algorithm in predicting SGA in our population was similar to that obtained in the reference population, according to sensitivity, but our false positive rate is significantly higher than the reference population.

Keywords:

• Clinical prediction model
• mass screening
• reproducibility of results
• sensitivity and specificity
• small for gestational age

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The present work was financially supported by Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro.