Biliary complications after liver transplantation: addition of T1-weighted images to MR cholangiopancreatography facilitates detection of cast in biliary cast syndrome
To evaluate the diagnostic performance of T2-weighted magnetic resonance (MR) cholangiopancreatography pulse sequences in comparison with MR cholangiopancreatography sequences combined with nonenhanced T1-weighted images in the detection of biliary cast syndrome in liver transplant recipients. This...
Gespeichert in:
Veröffentlicht in: | Radiology 2012-05, Vol.263 (2), p.429-436 |
---|---|
Hauptverfasser: | , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | To evaluate the diagnostic performance of T2-weighted magnetic resonance (MR) cholangiopancreatography pulse sequences in comparison with MR cholangiopancreatography sequences combined with nonenhanced T1-weighted images in the detection of biliary cast syndrome in liver transplant recipients.
This retrospective study was conducted in accordance with the declaration of Helsinki. Institutional review board approval was obtained. MR images in 95 patients who were examined after liver transplantation and who presented with symptoms of biliary obstruction were examined. Two separate sets of images, MR cholangiopancreatograms and MR cholangiopancreatograms plus T1-weighted images, were evaluated independently by three readers. Sensitivities, specificities, and positive and negative predictive values for biliary cast syndrome were calculated, and receiver operating characteristic curves were generated. The results of endoscopic retrograde cholangiopancreatography served as the reference standard. To determine interobserver agreement, κ values were calculated.
Cast appeared hyperintense on nonenhanced T1-weighted images. Sensitivities for T2-weighted MR cholangiopancreatography data alone were 0.65, 0.70, and 0.55 for the three readers. Adding unenhanced T1-weighted images resulted in sensitivities of 0.95, 0.90, and 0.90, respectively. Specificities for MR cholangiopancreatography alone and for MR cholangiopancreatography plus T1-weighted images were high on average (0.98, 0.97, and 0.97 vs 1.0 for all readers, respectively). Interobserver agreement was good for T2-weighted MR cholangiopancreatography (κ for readers 1 and 2 = 0.589, κ for readers 2 and 3 = 0.593, κ for readers 1 and 3 = 0.734) and was excellent for MR cholangiopancreatography plus T1-weighted images (κ for readers 1 and 2 = 0.806, κ for readers 2 and 3 = 0.881, κ for readers 1 and 3 = 0.848).
The combination of T2-weighted MR cholangiopancreatography and T1-weighted imaging yields higher diagnostic performance than MR cholangiopancreatography alone. Therefore, readers evaluating liver MR images with regard to biliary complications after liver transplantation should also look at the bile ducts on unenhanced T1-weighted images, as biliary cast might be more easily depicted on these images. |
---|---|
ISSN: | 0033-8419 1527-1315 |
DOI: | 10.1148/radiol.12111625 |