Key postnatal magnetic resonance characteristics for differentiating cystic biliary atresia from choledochal cyst.
Eur Radiol
; 2024 May 16.
Article
en En
| MEDLINE
| ID: mdl-38753194
ABSTRACT
OBJECTIVES:
To analyze the ability of magnetic resonance (MR) to identify cystic biliary atresia (CBA) and choledochal cyst (CC).METHODS:
Infants (≤ 1 year old) who were diagnosed with CBA or CC type I/IV from January 2010 to July 2023 were retrospectively reviewed. Imaging characteristics on MR were compared between the CBA and CC groups. Binary logistic regression and the area under the receiver operating characteristic curve (AUC) were analyzed for the identification of CBA.RESULTS:
Sixty-three patients with CBA (median age, 30 days) and 172 patients with CC (median age, 60 days) were included. Gallbladder (GB) wall thickness (cutoff, 1.2 mm) showed 98.4% sensitivity and 100% specificity (AUC, 0.998). MR-triangular cord thickness (MR-TCT) (cutoff, 4.1 mm) showed 100% sensitivity and 95.9% specificity (AUC, 0.986). The bile duct loop visualization showed 96.8% sensitivity and 100% specificity (AUC, 0.984). Proximal bile duct (PBD) diameter (cutoff, 1.3 mm) showed 92.1% sensitivity and 95.3% specificity (AUC, 0.977). Cyst wall thickness (cutoff, 1 mm) showed 77.8% sensitivity and 95.3% specificity (AUC, 0.942). The combination of GB wall thickness > 1.2 mm and MR-TCT > 4.1 mm, GB wall thickness > 1.2 mm and loop visualization, GB wall thickness > 1.2 mm, and cyst wall thickness > 1 mm showed 100% sensitivity and 100% specificity (AUC, 1.000).CONCLUSIONS:
Imaging characteristics on MR might be used to identify CBA and CC, and the combination of GB wall thickness and MR-TCT, or loop visualization, or cyst wall thickness, has a perfect diagnostic value. CLINICAL RELEVANCE STATEMENT Early and accurate differentiation of CBA and CC is essential, but current methods rely on inherently subjective ultrasound. Biliary features on MRI allow for an objective, accurate diagnosis.
Texto completo:
1
Banco de datos:
MEDLINE
Idioma:
En
Revista:
Eur Radiol
Asunto de la revista:
RADIOLOGIA
Año:
2024
Tipo del documento:
Article
País de afiliación:
China