Doppler ultrasound predictors of transplant hepatic venous outflow obstruction in pediatric patients.

OBJECTIVE

To investigate Doppler US and catheter venogram correlates to improve detection of transplant HVOO and avoid unnecessary invasive imaging procedures.

MATERIALS AND METHODS

A retrospective review was performed in all pediatric OLT patients undergoing catheter venography of the hepatic veins between 2007 and 2017 at a single large tertiary pediatric liver transplant institution.

RESULTS

Forty-four transplant hepatic venograms in 32 OLT patients were included (mean 1.38, range 1-4 venograms per patient). All venograms were preceded by an independent Doppler US examination. Twenty-one (47.7%) venograms were performed for the investigation of suspected HVOO based on Doppler US alone, 19 (43.2%) were performed for TJLB without suspected HVOO, 4 (9.1%) were performed for both. Sixteen (36.3%) instances of >50% anastomotic stenosis were identified. Mean peak anastomotic velocities were 208 cm/s and 116 cm/s in the presence and absence of a >50% venographic stenosis, respectively (P < 0.004). In all cases where there was a monophasic waveform seen on Doppler US, there was a > 50% stenosis seen on hepatic vein venogram. In all cases where a triphasic waveform was seen on Doppler US, there was no stenosis seen on hepatic vein venogram.

CONCLUSION

While a Doppler US velocity threshold providing both high sensitivity and specificity has yet to be identified, increasing peak anastomotic velocity and decreasing intrahepatic venous velocity correlate strongly with venographic outflow stenosis. The presence of a triphasic intrahepatic waveform provides good NPV.