Objective: Our purpose was to assess the value of sonographic measurement of fetal femur length in the second trimester, as a screening tool for Down's syndrome.
Study design: We evaluated a consecutive series of fetuses scanned by a single sonologist at the time of amniocentesis between 15 and 19 weeks. The study group consisted of fetuses with Down's syndrome (N = 16); the control group comprised normal fetuses (N = 1163). A linear regression model of the normal femur length based on biparietal diameter (BPD), was established for our population; the ratios of measured to expected femur length for a viven BPD (FL M/E) were calculated in the two groups. To test statistical significance of observed differences between case and control population, unpaired t test was used. The ability of specific FL M/E cut-off values to discriminate between Down syndrome and normal fetuses was assessed by Fisher's exact test.
Results: The mean ratio of measured to expected femur length was significantly lower in the Down syndrome as compared with control population (0.9473, DS 0.0795) versus 1.0, DS 0.0745) (p < 0.0045). The ratio of 0.91 or less predicted Down's syndrome with a sensitivity of 43.7% and a false-positive rate of 8.6%. For women with risk of one in 250 and one in 1000 of having an affected fetus based on maternal age, a shortened femur yielded positive predictive values of one in 26 and one in 105, respectively.
Conclusion: These results suggest that the sonographic measurement of fetal femur length for the screening of Down syndrome in the low-risk population is hindered by a high false positive rate (about 9%). It follows that the percentage of women requiring an amniocentesis would increase to un unacceptably high level. The utilization of this biometric marker may be helpful, in our opinion, for identifying fetuses at risk for Down syndrome in women between 35 and 38 years of age. These women in fact, are not offered amniocentesis for the prenatal diagnosis of Down's syndrome in the majority of italian institutions. The sonographic measurement of fetal femur length could detect about 45% of fetal Down's syndrome, offering an amniocentesis to 9% of women, with a consequent reduction of the cost required. Only a prospective study can evaluate the efficacy of this method to predict Down syndrome in such group of women.