Kang, X;
Cos Sanchez, T;
Arthurs, OJ;
Bevilacqua, E;
Cannie, MM;
Segers, V;
Lecomte, S;
... Jani, JC; + view all
(2019)
Postmortem fetal imaging: prospective blinded comparison of two‐dimensional ultrasound with magnetic resonance imaging.
Ultrasound in Obstetrics & Gynaecology
, 54
(6)
pp. 791-799.
10.1002/uog.20217.
Preview |
Text
Postmortem fetal imaging.pdf - Accepted Version Download (6MB) | Preview |
Abstract
OBJECTIVE: To compare the diagnostic rate and diagnostic accuracy of 3 Tesla (T) postmortem magnetic resonance imaging (PMMRI) and postmortem ultrasound (PMUS) in an unselected population. METHODS: In a blinded manner, we prospectively performed 3T PMMRI and PMUS on 160 unselected fetuses at 13-41 weeks of gestation. All imaging was reported according to a pre-specified template, for 5 anatomical regions: brain, thorax, heart, abdomen and spine. McNemar test for paired proportion was used to compare the non-diagnostic results rates of PMUS and PMMRI. When diagnostic, sensitivity, specificity and concordance rates for each anatomical region were calculated, using conventional autopsy as the reference standard. RESULTS: 3T PMMR performed significantly better than PMUS overall. Specifically, for the brain (7 non-diagnostic vs 43/160; 4.4% vs 26.9%; p<0.001), thorax (8 non-diagnostic vs 27/155; 5.2% vs 17.4%; p<0.001), heart (6 non-diagnostic vs 48/157; 3.8% vs 30.6%; p<0.001) and abdomen (5 non-diagnostic vs 37/157; 3.2% vs 23.6%; p<0.001). However, when diagnostic, we found similar accuracy for PMUS and PMMRI, with no difference in sensitivity or specificity, and similar concordance (PMUS 81.8-96.5%; PMMRI 81.6-99.1%). CONCLUSION: PMMRI performed significantly better than PMUS in this population, mainly due to a lower non-diagnostic study rate. PMMRI remains the first line imaging investigation for perinatal autopsy, but PMUS could be considered where MRI is not available, accepting a higher non-diagnostic rate. This article is protected by copyright. All rights reserved.
Archive Staff Only
View Item |