par Massez, Anne 
;Rypens, Françoise ;Metens, Thierry ;donner, catherine ;Avni, Efraim
Référence European radiology, 13, 9, page (2075-2081)
Publication Publié, 2003-09
;Rypens, Françoise ;Metens, Thierry ;donner, catherine ;Avni, Efraim Référence European radiology, 13, 9, page (2075-2081)
Publication Publié, 2003-09
Article révisé par les pairs
| Résumé : | The aim of this study was to evaluate the efficiency of the fetal iliac wing angle measurements in the detection of trisomy 21 during the second trimester and the impact of the fetal position on this measurement. During a 43-month period, the iliac wing angle was prospectively measured in 695 fetuses at genetic ultrasonography performed before amniocentesis. The iliac wing angle measurements were performed in a true axial section of the fetal pelvis and the relative position of the fetal spine was recorded (decubitus, lateral). Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and likelihood ratio (LR) were computed for multiple cutoff angles. The prevalence of trisomy 21 was 1.4%; 447 fetuses were in lateral position, 240 fetuses were in decubitus position, including, respectively, 2 and 8 fetuses with trisomy 21 (8 fetuses with other chromosomal anomalies were excluded). In euploid fetuses, the mean iliac wing angle was 83.7° in decubitus and 68.7° in lateral position; in fetuses with trisomy 21 the respective mean angles were 104.9° and 102.5°. A receiver operating characteristics (ROC) curve provided an absolute cutoff angle of 90°, predicting trisomy 21 with a sensitivity of 100%, a specificity of 72.4%, an NPV of 100%, a PPV value of 11.1%, and an LR of 3.6. The measurement of iliac wing angles in an axial section of the fetal pelvis is dependent on spine orientation and provides an efficient cutoff angle for the exclusion of trisomy 21 in patients at risk. |
