TY - JOUR
T1 - Doppler US for suspicion of hepatic arterial ischemia in orthotopically transplanted livers
T2 - Role of central versus intrahepatic waveform analysis
AU - Choi, Eugene K.
AU - Lu, David S.K.
AU - Park, Seong Ho
AU - Hong, Johnny C.
AU - Raman, Steven S.
AU - Ragavendra, Nagesh
PY - 2013/4
Y1 - 2013/4
N2 - Purpose: To compare the diagnostic performance of combinations of parameters derived from main hepatic artery (MHA) and intrahepatic artery (IHA) waveforms at Doppler ultrasonography (US), with the aim of developing a systematic approach to the evaluation of the hepatic arteries in orthotopic liver transplants in patients suspected of having hepatic arterial ischemia. Materials and Methods: This HIPAA-compliant retrospective study was approved by an institutional review board, with waiver of informed consent. From January 1, 2002, to November 1, 2011, 195 transplanted livers in 189 adults (129 men, 60 women; mean age, 53 years; age range, 18-73 years) who underwent Doppler US and follow-up (computed tomographic, magnetic resonance, or conventional) angiographic study within a 2-week interval were included. Diagnostic performance of the standard IHA and MHA criteria (resistive index [RI] < 0.5 and classic parvus tardus waveforms) with and without peak systolic velocity (PSV) thresholds (determined with receiver operating characteristic curve analysis) was assessed. The results of no-flow analysis and the most optimal MHA and IHA criteria were combined to create an algorithm, which was then applied to all liver transplants. Results: The standard criteria (RI < 0.5 and classic parvus tardus) demonstrated greater sensitivity (80% vs 55%, P = .008) when applied to IHA waveforms compared with MHA waveforms. Optimal PSV cutoff values were less than 67 cm/sec and 39 cm/sec for MHA and IHA, respectively. The addition of a PSV threshold resulted in significant decrease in overall accuracy when applied to IHA (87% vs 73%, P < .001) and MHA (82% vs 66%, P = .002) criteria. Application of an algorithm reflecting a combination of the most optimal MHA and IHA criteria and the results of no-flow analysis resulted in 96% sensitivity and 83% specificity. Conclusion: An algorithmic approach involving a tailored evaluation of the geographic distribution of absent flow and the quantitative parameters and waveform morphology of the MHA and IHAs allows for improved diagnostic performance in the detection of hepatic arterial complications in at-risk patients with orthotopic liver transplants.
AB - Purpose: To compare the diagnostic performance of combinations of parameters derived from main hepatic artery (MHA) and intrahepatic artery (IHA) waveforms at Doppler ultrasonography (US), with the aim of developing a systematic approach to the evaluation of the hepatic arteries in orthotopic liver transplants in patients suspected of having hepatic arterial ischemia. Materials and Methods: This HIPAA-compliant retrospective study was approved by an institutional review board, with waiver of informed consent. From January 1, 2002, to November 1, 2011, 195 transplanted livers in 189 adults (129 men, 60 women; mean age, 53 years; age range, 18-73 years) who underwent Doppler US and follow-up (computed tomographic, magnetic resonance, or conventional) angiographic study within a 2-week interval were included. Diagnostic performance of the standard IHA and MHA criteria (resistive index [RI] < 0.5 and classic parvus tardus waveforms) with and without peak systolic velocity (PSV) thresholds (determined with receiver operating characteristic curve analysis) was assessed. The results of no-flow analysis and the most optimal MHA and IHA criteria were combined to create an algorithm, which was then applied to all liver transplants. Results: The standard criteria (RI < 0.5 and classic parvus tardus) demonstrated greater sensitivity (80% vs 55%, P = .008) when applied to IHA waveforms compared with MHA waveforms. Optimal PSV cutoff values were less than 67 cm/sec and 39 cm/sec for MHA and IHA, respectively. The addition of a PSV threshold resulted in significant decrease in overall accuracy when applied to IHA (87% vs 73%, P < .001) and MHA (82% vs 66%, P = .002) criteria. Application of an algorithm reflecting a combination of the most optimal MHA and IHA criteria and the results of no-flow analysis resulted in 96% sensitivity and 83% specificity. Conclusion: An algorithmic approach involving a tailored evaluation of the geographic distribution of absent flow and the quantitative parameters and waveform morphology of the MHA and IHAs allows for improved diagnostic performance in the detection of hepatic arterial complications in at-risk patients with orthotopic liver transplants.
UR - http://www.scopus.com/inward/record.url?scp=84875420510&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84875420510&partnerID=8YFLogxK
U2 - 10.1148/radiol.12120557
DO - 10.1148/radiol.12120557
M3 - Article
C2 - 23297323
AN - SCOPUS:84875420510
SN - 0033-8419
VL - 267
SP - 276
EP - 284
JO - Radiology
JF - Radiology
IS - 1
ER -