Diagnostic accuracy of Doppler ultrasound for detecting hepatic artery stenosis after liver transplantation

Abstract

To evaluate the diagnostic accuracy of Doppler ultrasound (DUS) in detecting hepatic artery stenosis (HAS) after liver transplantation using computed tomography angiography (CTA) as the reference standard. This study included data from January 2005 to November 2017, where DUS of the hepatic artery of living and deceased donor liver grafts were compared with the reference standard, CTA. DUS parameters, such as intrahepatic artery (IHA) peak systolic velocity (PSV), resistive index (RI), systolic acceleration time (SAT); and extrahepatic artery (EHA) PSV were taken. The optimum cut-off was estimated using area under the receiver operating characteristic curve (AUC). Multivariable logistic regression analysis was developed to predict HAS. Ninety-nine liver transplant cases were retrieved, 50 met the inclusion criteria where nine patients had significant HAS. HAS patients had a significantly low IHAPSV with a cut-off of 35.1 cm/s (sensitivity 53.8%, specificity 78.4%, AUC 0.701). IHARI had a significantly low RI with a cut-off 0.585 (sensitivity 86.7%, specificity 85.4%, AUC 0.913). The IHASAT cut-off was 0.045 seconds (sensitivity 80%, specificity 91.4%, AUC 0.857). The EHAPSV cut-off was 197.4 cm/s (sensitivity 50%, specificity 99.1%, AUC 0.648). The prediction model using DUS parameters IHARI and IHASAT demonstrated good discrimination with an AUC of 0.930 (95% CI: 0.843, 1.000; sensitivity 93.3%, specificity 88%). A prediction model using the DUS parameters IHARI and IHASAT showed good diagnostic accuracy of 88.6% for monitoring liver transplant patients. If validated externally, this DUS model could be utilised to diagnose HAS in liver transplant recipients.