Vascular obstruction is a major complication after liver transplantation (LT). Hepatic artery stenosis is the most threatening complication because it can lead to loss of the grafted liver (3–8%) [1]. Clinical features in recipients with portal and hepatic vein obstructions vary according to the severity of the obstruction. Graft failure is an extreme event reserved for severe cases; and more commonly, vascular obstruction deteriorates postoperative course, by causing elevation of liver enzymes, ascites, abdominal distention, and gastrointestinal bleeding.

Ultrasonography is the initially accessible imaging technique for the evaluation of vascular complications in the early postoperative period and for long-term follow-up [2-4]. Doppler sonography is a quick, noninvasive, portable method for assessing the vasculature of a transplanted liver, and the examination can be performed at the patient bedside. Several Doppler parameters are feasible for detecting vascular complications. Recently, the use of contrast-enhanced ultrasound (CEUS) has been increased to evaluate the LT patients. The use of ultrasound contrast agent (USCA) makes grayscale contrast enhancement image similar to that of contrast– enhanced computed tomography (CT) which enables detection of parenchymal perfusion defects, ischemic changes and vascular congestion. Moreover, it boosts Doppler signals and may aid in evaluating the Doppler spectrum in hepatic vasculatures having weak or otherwise undetectable Doppler signals [5].

The signals from conventional Doppler examination are reflected from red blood cells [6]. After administering USCA, the signals are enhanced by the microbubbles. The serial follow-up of Doppler parameters in LT patients is important for the diagnosis of vascular complications (such as hepatic artery steal syndrome and portal steal syndrome) and acute cellular rejection. If Doppler parameters are affected by the administration of USCA on serial follow up, the occurrence and deterioration of liver pathology can be masked in postoperative Doppler examination for LT patients using USCA. Clinical validation is required to determine whether Doppler measurements are comparable before and after administering USCA. This study aimed to compare the spectral Doppler velocities and parameters that were measured before and after administering USCA in an immediate postoperative period for recipients of LT.

A summary of the results is presented in Table 1. Before administration of USCA, the mean peak systolic and end-diastolic velocities of the graft hepatic artery were 79.0 cm/s (SD, ±45.2 cm/s; range, 20.1–222.0 cm/s) and 23.0 cm/s (SD, ±19.5 cm/s; range, 2.5–109.0 cm/s), respectively. The mean systolic acceleration time and resistive index were 65.3 ms (SD, ±14.4 ms; range, 33–100 ms) and 0.71 (SD, ±0.12; range, 0.47–1.01), respectively. The mean peak flow velocity of the graft portal vein was 74.3 cm/s (SD, ±35.4 cm/s; range, 13.9–193.0 cm/s). The mean peak flow velocity and venous pulsatility index of the graft hepatic vein were 47.0 cm/s (SD, ±22.6 cm/s; range, 13.7–118.0 cm/s) and 0.65 (SD, ±0.41; range, 0.1–1.7), respectively.

After USCA injection, 114–287 s (187.6±37.1 s) were spent evaluating vascular structures and parenchymal perfusion using low mechanical index grayscale harmonic imaging. Spectral Doppler measurement was thereafter initiated.

After administration of USCA, both peak systolic and end-diastolic velocities increased in 22 (61%) patients and decreased in 14 (39%) patients compared with the velocities obtained before USCA injection. The mean peak systolic and mean end-diastolic velocities of the graft hepatic artery after administering USCA were 89.0 cm/s (SD, ±56.0 cm/s; range, 22.9–312.0 cm/s) and 25.2 cm/s (SD, ±22.1 cm/s; range, 3.3–124.0 cm/s), respectively. Both values increased on an average by 13% and 10%, respectively, compared with values that were obtained before USCA injection,; however, the increments were not statistically significant (P=0.097 and 0.103, respectively) (Fig. 1A).

Both the systolic acceleration time and resistive index of the graft hepatic artery increased in 23 (64%) patients and decreased in 13 (36%) after administering USCA. The mean systolic acceleration time and resistive index after administration of USCA were 67.6 ms (SD, ±19.7 ms; range, 42–116 ms) and 0.72 (SD, ±0.11; range, 0.54–0.93). The systolic acceleration time increased on an average by 4%, and the resistive index increased by 1% after administering USCA, but neither were statistically significant differences (P=0.489 and 0.205, respectively) (Fig. 1B).

The peak flow velocity of the graft portal vein increased in 22 (61%) patients and decreased in 14 (39%) after administering USCA. The mean peak flow velocity after administering USCA was 82.0 cm/s (SD, ±39.3 cm/s; range, 10.7–181.0 cm/s) showing 10% increase compared with that obtained before USCA injection. The difference was not statistically significant (P=0.128) (Fig. 1C). The peak flow velocity of the graft hepatic vein increased in 25 (69%) patients and decreased in 11 (31%) after administering USCA. The mean peak flow velocity of the graft hepatic vein after administering USCA was 53.3 cm/s (SD, ±27.8 cm/s: range, 16.4–122.0 cm/s), which is a 13% increase compared with that obtained before USCA injection. The difference was not statistically significant (P=0.190) (Fig. 1D).

The venous pulsatility index of the graft hepatic vein increased in 19 (53%) patients and decreased in 17 (47%) after administration of USCA. The mean venous pulsatility index after administering USCA was 0.68 (SD, ±0.41; range, 0.13–1.48), showing 5% increase compared with the value before administering USCA. The difference was not statistically significant (P=0.494) (Fig. 1D).

We found that although there was a tendency for Doppler velocities in the hepatic artery and portal and hepatic veins to increase after administering USCA, with mean increases of 10–13%, they did not significantly differ from the values that were obtained before USCA. And the calculated Doppler parameters, the resistive index of the graft hepatic artery and venous pulsatility index of graft hepatic vein, were not significantly affected by administering USCA (1% and 5% increments, respectively) nor was the graft hepatic artery systolic acceleration time significantly affected by administering USCA (4% increment).

Whether Doppler velocities change after administering USCA is a controversial topic. According to Forsberg, et al. [7], after administration of USCA, Doppler velocities increased from 20% to 45% in a rabbit aorta model and 17% in a flow phantom. They supposed that the observed increments in peak velocity measurements represented an artifact that was produced by contrast agents, and this phenomenon was probably related to a strong enhancement of the signal intensity exceeding the dynamic range of their scanner. In a cardiac model reported by Yokoyama, et al. [8], the Doppler velocity increased by 13% after administration of USCA. In contrast, other experimental and clinical studies have demonstrated that Doppler velocities were not significantly changed after administering USCA with adequate adjustment of the Doppler gain [9,10]. Most of the previous studies were conducted using phantom or animal models or evaluated large superficial vessels or intracardiac flow in humans. In this study, we attempted to determine whether measuring Doppler velocities and other parameters is significantly affected by the use of USCA in common clinical settings of postoperative evaluation of recipients of LT.

In the current study, the ranges of the Doppler velocity increments that we documented were wider (–61~270%) than those documented in previous studies. In addition, we frequently encountered a decrease in the Doppler velocities in some patients after administration of USCA, unlike in previous studies. We assumed that the wide range of change in velocity and the finding of velocity decrease in some patients after the use of USCA might have been caused by several confounding factors other than USCA affecting the spectral Doppler velocity measurement. In the clinical setting of performing hepatic Doppler after LT, it may sometimes be difficult to assess the exact direction of blood flow. This difficulty can be encountered, particularly in the case of a small tortuous hepatic artery that may not be traceable on a grayscale image and only discontinuously appears on color flow imaging. Furthermore, if there is a helical or turbulent flow, which is common in the graft portal vein when there is a donor–recipient size discrepancy or mild anastomotic stenosis, the blood flow may not be parallel to the longitudinal axis of the vessels. Also, although Doppler velocity measurements before and after the administration of USCA were performed by the same radiologist in our study, there may have been minute differences in the adequacy of Doppler angle setting between the two examinations. Lastly, exaggerated respiratory motion may have disturbed the stability of Doppler velocity measurement.

In our study, we did not perform the Doppler examination immediately after USCA administration. There was a time lag of 114–287 s, depending on the time spent for the harmonic grayscale enhancement imaging to evaluate the graft vessels and parenchyma. Therefore, a variable degree of metabolic and mechanical breakage of the contrast microbubbles may have occurred before we initiated Doppler measurements, which may have changed the USCA concentration to unknown degree. However, this is unlikely to have caused serious variation in the USCA concentration because the period beginning from 90 s after administering the contrast is hemodynamically classified as an equilibrium phase following a single bolus intravenous injection of the contrast agent [11]. And, the low mechanical index used for grayscale harmonic imaging would be unlikely to cause much destruction of microbubbles. Nevertheless, the effects of time interval after USCA administration to the changes of USCA concentration and the Doppler frequency shift are not evaluated before, and a further study with strictly controlled scan time after administration of USCA may be meaningful.

There are some limitations in our study. The first is the variable time elapse before post-contrast Doppler velocity measurement as described above. The second is that we did not evaluate inter-observer variability. Doppler assessment is a repeatable method for measuring hepatic blood flow, with stable intra-observer variability [12]. However, inter-observer variability of Doppler measurements has been reported to range from 0% to 10% [13,14]. In the usual clinical setting, postoperative serial follow-up Doppler examinations of recipients of LT are not performed by one examiner, and therefore, Doppler parameters before and after administering USCA could be affected by inter-observer variability. However, all Doppler examinations in our study were performed by one examiner, and the inter-observer variability was not considered.

In conclusion, the measured velocities of graft hepatic vessels tended to increase after administration of USCA in recipients of LT; however, the differences were not statistically significant. The calculated Doppler parameters (resistive index of hepatic artery and venous pulsatility index of hepatic vein) and the systolic acceleration time of the graft hepatic artery were not significantly affected by administration of USCA. Therefore, the comparison of these parameters between the exams performed with or without administering USCA is valid in Doppler surveillance of recipients of LT.