Reproducibility of measurements of hepatic artery and portal vein diameter and flow velocity in paediatricliver transplant recipients

Hepatic blood flow is critical for successful liver transplantation. To reliably calculate flow, vessel cross-sectional area and mean flow velocities must be reproducibly measurable. To assess the reproducibility of sonographic measurements of portal vein (PV) and hepatic artery (HA) diameters and mean flow velocities in children after transplantation. Ten children were scanned by two operators 5 or 7 days after transplantation. Each scanned the same patient twice measuring the diameter and the time-averaged velocity (TAV) in the PV and HA. Operators were blinded to all measurements. Agreement and repeatability were analysed statistically. The interobserver variation in PV and HA diameters and TAVs were expressed as mean differences between operators and plotted against the mean value for each. This showed relatively poor agreement for PV diameter (mean difference 1.23 +/- 0.96 mm), but better agreement for HA diameter (mean difference 0.18 +/- 0.6 mm). Intraobserver differences were of similar magnitude. TAVs showed better agreement and repeatability. This study demonstrates substantial inter- and intraobserver variation in sonographic measurements of vessel diameters and TAV after paediatric liver transplantation. Estimates of volume of blood flow to the transplant derived from these measurements are poorly reproducible.