Predicting down syndrome: a comparative evaluation of nasal bone length in mid-trimester pregnancy

BACKGROUND: This study aimed to evaluate the diagnostic performance of fetal nasal bone length (NBL), the biparietal diameter to nasal bone length ratio (BPD/NBL), and nasal bone length multiples of the median (NBL-MoM) values for Down syndrome (DS) screening between 17 and 25 weeks of gestation in a population-specific cohort. METHODS: A retrospective analysis was conducted on fetuses between 17 and 25 gestational weeks. Normative data for NBL and NBL-MoM were generated using gestational age–specific reference medians, and regression equations were established. The NBL, BPD/NBL ratio, and NBL-MoM values were compared between DS and euploid groups. NBL-MoM values were calculated using raw (non-transformed) measurements divided by the expected gestational age–specific median, and euploid status was confirmed by normal postnatal outcomes. Sensitivity, specificity, and likelihood ratios were calculated for various cutoff points. RESULTS: NBL showed a significant linear correlation with gestational week (R² = 0.31, p < 0.001), while BPD/NBL was independent of gestational age. In DS cases, absence of the nasal bone was observed in 40% of fetuses, compared to 0.03% in controls (sensitivity: 40%, specificity: 100%, LR⁺: 1228). An NBL < 2.5 mm yielded a sensitivity of 49% and specificity of 99.9% (LR⁺: 746). Using the 5th percentile as a cutoff, sensitivity and specificity reached 86% and 98.9%, respectively (LR⁺: 75, LR⁻: 0.15). ABPD/NBL ratio ≥ 9 yielded 93% sensitivity and 97.5% specificity (LR⁺: 37). NBL-MoM values < 0.65 yielded 97% sensitivity and 98% specificity. CONCLUSION: Our findings support the use of population-specific reference ranges for fetal NBL, BPD/NBL ratio, and NBL-MoM values in the second trimester to improve the screening accuracy for Down syndrome. These sonographic markers—particularly NBL < 5th percentile, BPD/NBL ≥ 9, and NBL-MoM < 0.65—may provide valuable, non-invasive indicators to support clinical assessment. TRIAL REGISTRATION: This study was retrospectively registered and authorized by the local Ethics Committee of Ankara University (approval number: 2025/696; date: 03 September 2025).