Accuracy of single ultrasound parameters in detection of fetal growth restriction

Abstract

The objective of this study was to evaluate the diagnostic accuracy of different single ultrasonographic parameters in predicting intrauterine growth-restricted (IUGR) babies as defined by Ponderal index at birth. Study sample composed of two sets of data from Iran and Australia. The Iranian sample consisted of 296 Iranian women. All the study women received prenatal care and delivered at Fatemieh Hospital in Shahrood, Iran. The data from 219 Australian fetuses were obtained from the ultrasound section at the Wollongong Hospital in Australia. Ultrasonographic measurements of biparietal diameter (BPD), femur length (FL), head circumference (HC), abdominal circumference (AC), amniotic fluid index (AFI), and Doppler from umbilical arteries (S/D ratio) were obtained. Only those pregnancies were included in which the estimated date of delivery (EDD) by LMP (last menstrual period) agreed within 14 days with the estimated date of delivery determined by the initial ultrasound examination. Sensitivity (SE), specificity (SP), positive predictive value (PPV) and negative predictive value (NPV) were calculated for single proposed ultrasound parameters in the Iranian and Australian samples. When different variables are compared, with a cut-off point at or below the 10th percentile AC and HC had the highest sensitivities in the Australian sample while AC was the most sensitive parameter for IUGR detection in the Iranian sample. BPD has a reasonably high sensitivity at this threshold. The AFI ratio has the lowest sensitivity in predicting IUGR in the Australian sample. Positive predictive values were low in all of the parameters in both the Iranian and Australian samples. By increasing the cut-off for PI to the 20th percentile, the sensitivity of reduced AC increased in the Australian sample while the sensitivity for an abnormal S/D ratio and reduced AFI to identify IUGR decreased. Our results indicate that reduced AC was the best single parameter in discriminating between IUGR and non-IUGR fetuses with the highest sensitivity among the proposed parameters in the both Iranian and Australian sample. However, the PPV of this parameter is low. This means that a high number of false-positive cases are detected using each parameter, which reduces the usefulness of identification. Other ultrasound obstetrical parameters may also have a reasonable level of sensitivity, however, the PPV of all parameters is low. On the whole, our results show that although the examined ultrasonographic criteria may detect a group of fetuses that need close antepartum surveillance, none of these parameters are appropriate enough to be used in isolation in clinical practice. Using single ultrasound parameters does not have high sensitivity and PPVs in detection of fetal growth restriction. This limits accuracy and utility of these tests in the detection of IUGR fetuses.