OBJECTIVES: Examination of the fetal nasal bones by ultrasound between 11 and 14 weeks of gestation has been proposed as an additional tool in the detection of trisomy 21 in a high-risk population. However the variability in the identification of fetal nasal bones by ultrasound has not yet been investigated. The aim of this study was to assess the intraobserver and interobserver reproducibility of fetal nasal bone identification by ultrasound between 11 and 14 weeks of gestation. METHODS: A total of 1040 consecutive ultrasound examinations were performed at 11-14 weeks of gestation for nuchal translucency (NT) measurement and nasal bone identification by ultrasound. A total of 657 consecutive video-loops were assessed by three experienced operators. Each operator assigned cases to one of three categories, namely present, uncertain or absent nasal bones, and the results were compared between operators. To assess the intraoperator variability each operator reviewed 100 randomly selected videos out of the 657 loops and again used the same classification. Results were compared by pairwise unweighted and weighted Kappa index to evaluate the inter- and intraoperator variability. RESULTS: Among the 1040 fetuses, there were 51 (4.9%) with an NT measurement above the 95th centile. Nasal bones were identified by ultrasound in 948, not seen in eight and impossible to assess in 84 fetuses. Four fetuses had trisomy 21 including three with absent nasal bones and increased NT and one with present nasal bones and normal NT. The Kappa and weighted Kappa values for interoperator variability between the three operators were between 0.26 and 0.37 and 0.33 and 0.44, respectively. The Kappa and weighted Kappa values for intraoperator variability were between 0.35 and 0.48 and 0.43 and 0.55, respectively. CONCLUSION: The assessment of fetal nasal bones is only fairly reproducible. Although the performance of the test in fetuses at high risk for trisomy 21 has been reported to be good, its implementation as an additional screening technique in the general population must be accompanied by teaching and quality control programs. Copyright 2003 ISUOG. Published by John Wiley & Sons, Ltd.
OBJECTIVES: Examination of the fetal nasal bones by ultrasound between 11 and 14 weeks of gestation has been proposed as an additional tool in the detection of trisomy 21 in a high-risk population. However the variability in the identification of fetal nasal bones by ultrasound has not yet been investigated. The aim of this study was to assess the intraobserver and interobserver reproducibility of fetal nasal bone identification by ultrasound between 11 and 14 weeks of gestation. METHODS: A total of 1040 consecutive ultrasound examinations were performed at 11-14 weeks of gestation for nuchal translucency (NT) measurement and nasal bone identification by ultrasound. A total of 657 consecutive video-loops were assessed by three experienced operators. Each operator assigned cases to one of three categories, namely present, uncertain or absent nasal bones, and the results were compared between operators. To assess the intraoperator variability each operator reviewed 100 randomly selected videos out of the 657 loops and again used the same classification. Results were compared by pairwise unweighted and weighted Kappa index to evaluate the inter- and intraoperator variability. RESULTS: Among the 1040 fetuses, there were 51 (4.9%) with an NT measurement above the 95th centile. Nasal bones were identified by ultrasound in 948, not seen in eight and impossible to assess in 84 fetuses. Four fetuses had trisomy 21 including three with absent nasal bones and increased NT and one with present nasal bones and normal NT. The Kappa and weighted Kappa values for interoperator variability between the three operators were between 0.26 and 0.37 and 0.33 and 0.44, respectively. The Kappa and weighted Kappa values for intraoperator variability were between 0.35 and 0.48 and 0.43 and 0.55, respectively. CONCLUSION: The assessment of fetal nasal bones is only fairly reproducible. Although the performance of the test in fetuses at high risk for trisomy 21 has been reported to be good, its implementation as an additional screening technique in the general population must be accompanied by teaching and quality control programs. Copyright 2003 ISUOG. Published by John Wiley & Sons, Ltd.